Wartime Commercial Policy
Wartime Commercial
Policy and Trade
between Enemies
Mariya Grinberg
Conventional wisdom
suggests that trade is the ªrst casualty of war.1 Because the gains from trade
can be converted into military capabilities, trading with the enemy is akin to
selling the opponent the gun they will use to shoot you. The empirical record
of wartime trade, Tuttavia, suggests otherwise. Per esempio, World War I, UN
total war in which the majority of the states involved fought for their very sur-
vival, saw extensive trade between enemy belligerents. Britain continued to
trade with its enemies until October 1, 1918—one month and eleven days be-
fore the Armistice. Infatti, Britain started the war with restrictions on the ex-
port of only 20 percent of the goods that it ultimately prohibited from reaching
the enemy. Even after a year of ªghting, by the end of August 1915, around
half of the products that would eventually be prohibited were still allowed to
be legally traded with enemy states.
World War I is hardly unique in that trade occurred and varied during the
war. Some enemies continue to trade throughout the war—for example, India
and Pakistan in the First Kashmir War (1947–49) and Yugoslavia and Croatia in
the War of Bosnian Independence (1992).2 Other states sever trade immedi-
Mariya Grinberg is an assistant professor at the Massachusetts Institute of Technology.
The author is grateful for comments and suggestions from Anjali Anand, Nick Anderson, Eliza
Gheorghe, Kelly Greenhill, Robert Gulotty, Daniel Jacobs, Tyler Jost, John Mearsheimer, Asfandyar
Mir, Paul Poast, Kevin Weng, and the anonymous reviewers. The author also appreciates the valu-
able feedback offered by participants at the Workshop on International Politics at the University of
Chicago, the Belfer Center for Science and International Affairs at the Harvard Kennedy School,
and the Project on the Political Economy of Security at Boston University. A novel dataset is avail-
able in the online appendix at doi.org/10.7910/DVN/ZLM9TV.
1. Dale C. Copeland, Economic Interdependence and War (Princeton, N.J.: Princeton University
Press, 2015); Ronald Findlay and Kevin H. O’Rourke, Power and Plenty: Trade, War, and the World
Economy in the Second Millennium (Princeton, N.J.: Princeton University Press, 2007); Reuven Glick
and Alan M. Taylor, “Collateral Damage: Trade Disruption and the Economic Impact of War,"
Working Paper (Cambridge, Massa.: National Bureau of Economic Research, May 2008 [agosto
2005]), http://www.nber.org/papers/w11565; Charles H. Anderton and John R. Carter, “The Im-
pact of War on Trade: An Interrupted Times-Series Study,” Journal of Peace Research, Vol. 38, No. 4
(Luglio 2001), pag. 445–457, doi.org/10.1177/0022343301038004003; Katherine Barbieri and Jack S.
Levy, “Sleeping with the Enemy: The Impact of War on Trade,” Journal of Peace Research, Vol. 36,
No. 4 (Luglio 1999), pag. 463–479, doi.org/10.1177/0022343399036004005; Joanne Gowa, Allies, Adver-
saries, and International Trade (Princeton, N.J.: Princeton University Press, 1995); Richard N.
Rosecrance, The Rise of the Trading State: Commerce and Conquest in the Modern World (New York: Ba-
sic Books, 1987); Solomon William Polachek, “Conºict and Trade,” Journal of Conºict Resolution,
Vol. 24, No. 1 (Marzo 1980), pag. 55–78, doi.org/10.1177/002200278002400103; and Kenneth N.
Waltz, Theory of International Politics (Long Grove, Ill.: Waveland, 1979).
2. On the First Kashmir War, see Syed Mansoob Murshed, Hugh Ward, and Han Dorussen, “Any
International Security, Vol. 46, No. 1 (Estate 2021), pag. 9–52, https://doi.org/10.1162/isec_a_00412
© 2021 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
9
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 10
ately at the start of the war—for example, England and Argentina in the
Falkland Islands War (1982) and India and Pakistan in the Kargil War (1999).3
Yet, other states start off trading with the enemy, only to change course during
the war, as occurred, Per esempio, between Ethiopia and Somalia in the Second
Ogaden War (1977–78).4 There is remarkable variation in wartime trading pat-
terns between adversaries.
Why do states trade with their enemies in wartime? In questo articolo, I argue
that states make deliberate choices when setting their wartime commercial
policies and that these policies are tailored to the type of war the state expects
to ªght. Speciªcally, states seek to balance two goals—maximizing revenue
from continued trade during the war and minimizing the opponent’s ability to
beneªt militarily from trade.
Di conseguenza, states have two reasons to continue trading with their enemies
during war. Primo, states continue to trade in products that their opponents take
a long time to convert into military capabilities, because the security conse-
quences from this trade will not accrue in time to help the opponent win the
war. Secondo, states continue to trade in products that are essential to the do-
mestic economy but that can be obtained only from the opponent, because
sacriªcing this trade would impair the state’s long-term security. Inoltre,
states revise their wartime commercial policies based on how well they per-
form on the battleªeld. As the expected length of a war increases, the number
of prohibited products will increase, because the opponent will have more
Ties that Bind? Economic Diplomacy on the South Asian Subcontinent,” Hague Journal of Diplo-
macy, Vol. 6, Nos. 1–2 (2011), pag. 149–169, doi.org/10.1163/187119111X569046; and Zareen Fatima
Naqvi, Philip Schuler, and Kaspar Richter, “Pakistan-India Trade: Overview and Key Issues,"
in Naqvi and Schuler, eds., The Challenges and Potential of Pakistan-India Trade (Washington, D.C.:
World Bank, 2007), pag. 1–28, http://documents1.worldbank.org/curated/en/990861468074627
673/pdf/402730P07493901India1Trade01PUBLIC1.pdf. On the War of Bosnian Independence, Vedere
Peter Andreas, “The Clandestine Political Economy of War and Peace in Bosnia,” International
Studies Quarterly, Vol. 48, No. 1 (Marzo 2004), pag. 29–51, https://www.jstor.org/stable/3693562;
and John Mueller, The Remnants of War (Ithaca, N.Y.: Cornell University Press, 2013).
3. On the Falkland War, see Lisa L. Martin, “Institutions and Cooperation: Sanctions during the
Falkland Islands Conºict,” International Security, Vol. 16, No. 4 (Primavera 1992), pag. 143–178, doi.org/
10.2307/2539190; and M.S. Daoudi and M.S. Dajani, “Sanctions: The Falklands Episode,” World To-
day, Vol. 39, No. 4 (April 1983), pag. 150–160, https://www.jstor.org/stable/40395502. On the Kargil
War, see Rashid Amjad and Shahid Javed Burki, eds., Pakistan: Moving the Economy Forward (Camera-
ponte: Cambridge University Press, 2015); and Michael Kugelman and Robert M. Hathaway, eds.,
Pakistan-India Trade: What Needs to Be Done? What Does It Matter? (Washington, D.C.: Woodrow Wil-
son International Center for Scholars, Asia Program, 2013).
4. This can be seen in the dyadic trade data from Cow Trade v 4. Katherine Barbieri, Omar M.G.
Keshk, and Brian M. Pollins, “Trading Data: Evaluating Our Assumptions and Coding Rules,"
Conºict Management and Peace Science, Vol. 26, No. 5 (novembre 2009), pag. 471–491, doi.org/
10.1177/0738894209343887.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 11
time to beneªt militarily from the gains of trade. Allo stesso modo, the closer the war
is to becoming an existential threat, the greater the portion of wartime trade
with the enemy that the state will sever.
The article makes two major theoretical contributions. Primo, it shows that
temporality is key to understanding the security externalities of trade—that is,
the military consequences of a state beneªting from trade. Existing scholarship
focuses on the idea that trading with the enemy increases the adversary’s mili-
tary capabilities,5 but it omits the temporal dimension, in which economic
gains may be converted into military power. Although all gains from trade
are ultimately convertible into military capabilities, the amount of time this
process takes varies by product. A similar temporal distinction can be applied
to all wartime policy tools to determine if, and to what extent, they carry secu-
rity externalities.
Secondo, the article challenges a central conclusion of economic inter-
dependence theory—that signiªcantly interdependent states are least likely to
ªght each other. According to that theory, trade between states is severed
in war, which incentivizes states to avoid war to prevent losing the beneªts
of trade.6 So long as trade is lost during war, trade deters conºict. As my
research shows, Tuttavia, under the right circumstances, states have ample
reason to trade with their enemies during war. Additionally, the more interde-
pendent two economies are, the greater their incentives for wartime trade.
Contrary to the predictions of economic interdependence theory, trade is un-
likely to serve as a deterrent to war between highly interdependent states. Questo
ªnding is particularly salient given the heightened possibility of conºict
between the United States and China. Although they share signiªcant eco-
nomic ties, these ties could not be used to prevent the escalation of a potential
U.S.-China conºict.
5. Gowa, Allies, Adversaries, and International Trade; Joanne Gowa and Edward D. Mansªeld,
“Power Politics and International Trade,” American Political Science Review, Vol. 87, No. 2 (Giugno
1993), pag. 408–420, doi.org/10.2307/2939050; and Thomas C. Schelling, International Economics
(Boston: Allyn and Bacon, 1958).
6. For an excellent review of this literature, see Copeland, Economic Interdependence and War;
Edward D. Mansªeld and Brian M. Pollins, “The Study of Interdependence and Conºict: Recente
Advances, Open Questions, and Directions for Future Research,” Journal of Conºict Resolution,
Vol. 45, No. 6 (Dicembre 2001), pag. 834–859, https://www.jstor.org/stable/3176160; and Susan M.
McMillan, “Interdependence and Conºict,” Mershon International Studies Review, Vol. 41, No. 1
(May 1997), pag. 33–58, doi.org/10.2307/222802. For an alternative point of view, see Waltz, Theory
of International Politics; and Robert Gilpin, “Economic Interdependence and National Security in
Historical Perspective,” in Klaus Knorr and Frank N. Trager, eds., Economic Issues and National Se-
curity (Lawrence: University Press of Kansas, 1978), pag. 19–66.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 12
The article proceeds in ªve sections. The ªrst section explains what
constitutes trading with the enemy. The second describes the existing literature
that deals with the phenomenon. The third provides a theory explaining why
warring states continue to trade with each other. In the fourth section, I discuss
the research design. The ªfth section tests the theory by drawing evidence
from Britain’s wartime commercial policy throughout World War I. IL
conclusion offers some policy implications for possible wartime trade in fu-
ture wars.
What Is Wartime Commercial Policy?
The main outcome of interest in this article is a state’s wartime commercial
policy toward an enemy belligerent. It is deªned as the collection of decisions
that a state makes about which products of enemy origin the state allows to be
imported during a given war, and which products it permits to be exported to
the enemy during that war. As the term suggests, a wartime commercial policy
is applicable only during war. Once a country enters a state of war, the existing
level of trade between the two enemy belligerents becomes governed by the
country’s wartime commercial policy. Unlike peacetime, when a state can af-
ford to include a wide range of concerns and interests in its commercial deci-
sions, during war, every decision about the enemy state has to be judged based
on its potential impact on a battleªeld victory. The security motivations for
severing trade with the enemy are highest during war, thus presenting the
hardest conditions for trade with the enemy to exist.
Wartime commercial policy refers to economic cooperation between enemy
belligerents. It excludes trade between belligerents and their allies. Likewise, Esso
excludes trade between neutral states and the belligerents in products of neu-
tral origin. Per esempio, trade between the United States and Germany in 1940
is not trade with the enemy, as the two countries were not yet enemy bellig-
erents. In contrasto, trade in 1940 between Sweden and Germany in products of
British origin was part of the Anglo-German trade with the enemy. Lastly, war-
time commercial policy distinguishes between legal and prohibited trade. For
instance, it excludes contraband trade, as such trade stands in direct opposi-
tion to the preferences of the state.
Trade between enemy belligerents can be direct or indirect. Direct trade re-
fers to an exchange of merchandise in which ownership of the products
is transferred from a merchant in state A directly to a merchant in state B.
When the ownership of a product switches to an intermediary before being de-
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 13
livered to its ªnal destination, the trade is considered indirect. Indirect trade
requires the participation of neutral states; Tuttavia, only trade in products of
belligerent origin is relevant. The simplest form of this is the sale of goods
from a merchant in state A to a merchant in a neutral state, which then sells the
same goods to a merchant in state B. Indirect trade also includes cases of do-
mestic substitution in a neutral country. Per esempio, a merchant in a neutral
state imports a product from state A, then sells an identical product that was
manufactured in the neutral state to a merchant in state B. Another form of in-
direct trade is minimal manufacturing, which requires a merchant in a neutral
country to import a raw material or an intermediate good from state A, pro-
cess this good in some minimal manner, then export the ªnished product to
state B.
When states are making wartime commercial policies, they set limitations on
both direct and indirect trade with the enemy. Così, there is variation in war-
time commercial policies between allowing all direct and indirect trade, vari-
ous restrictions on direct trade and various restrictions on indirect trade, E
no direct or indirect trade with the enemy.
To isolate the reasons why a state would trade with the enemy, I theorize a
state’s preferences for wartime trade. The theory does not deal with the ulti-
mate level of bilateral trade between enemy belligerents. Primo, bilateral trade
frequently excludes indirect trade. Secondo, and more importantly, the bilateral
level of trade is affected by multiple processes beyond the preferences of
belligerents for wartime trade—for example, the progress of war destroying
certain trade routes, the increased risk of trade and availability of insurance af-
fecting merchants’ willingness to engage in trade, and the infrastructural logis-
tics of open ports limiting the amount of trade that can be processed. Looking
at a state’s wartime commercial policy, as opposed to bilateral trade, Perciò
focuses the analysis on the maximum amount of trade a state allows during a
war, isolating how a state resolves the dilemma between economic and secu-
rity incentives without confusing the analysis with other factors that affect the
observed level of bilateral trade.
Allo stesso modo, the analysis does not focus on the decisions of individual ªrms to
engage in trade with the enemy during war. While in many economies it is the
ªrms that make decisions about what to trade in, with whom, and how much,
it is the state that sets the limits within which the ªrms are bound to operate.
To the extent that the ªrms operate within these boundaries, they are driven
mostly by proªt maximization, making a ªrm-level decision to trade with
the enemy less surprising. The state does not have the same luxury, espe-
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 14
cially in wartime. States have to balance economic motivations with secu-
rity concerns, which makes the decision to trade with the enemy considerably
more puzzling.
Existing Explanations of Wartime Trade
There are very few attempts in the literature to explain why belligerents
continue to trade with their enemies during war. Possible explanations for
wartime trade can be drawn from three bodies of literature that seek to con-
nect security and economic concerns: economic interdependence, relative
gains, and domestic political economy. The arguments made in these ªelds do
non, Tuttavia, directly deal with wartime commercial policy, focusing instead
on peacetime.7 Problematically, when these theories are expanded to wartime,
they tend to provide an absolute prediction for all wars—either trade with the
enemy should never occur, or it should always occur.
Economic interdependence theory argues that the more states trade with
each other, the less likely they are to ªght. Three causal mechanisms could ex-
plain this pattern; all three depend on the assumption that trade between
belligerents is severed during war. By far the most prevalent causal mecha-
nism is opportunity cost.8 The opportunity cost of ªghting is the value of trade
7. The notable exception is the 2004 work of Jack S. Levy and Katherine Barbieri. They explicitly
do not claim to offer a theory to explain wartime trade, but provide numerous insightful hypothe-
ses that are discussed later in the article. Levy and Barbieri, “Trading with the Enemy during War-
time,” Security Studies, Vol. 13, No. 3 (2004), pag. 1–47, doi.org/10.1080/09636410490914059. Vedere
also Barbieri and Levy, “The Trade-Disruption Hypothesis and the Liberal Economic Theory of
Peace,” in Gerald Schneider, Barbieri, and Nils Petter Gleditsch, eds., Globalization and Armed
Conºict (Lanham, Md.: Rowman and Littleªeld, 2003), pag. 277–298; and Barbieri and Levy,
“Sleeping with the Enemy.”
8. John R. Oneal and Bruce M. Russett, “The Classical Liberals Were Right: Democracy, Interde-
pendence, and Conºict, 1950–1985,” International Studies Quarterly, Vol. 41, No. 2 (Giugno 1997),
pag. 267–293, https://www.jstor.org/stable/3013934; Katherine Barbieri, The Liberal Illusion: Fa
Trade Promote Peace? (Ann Arbor: University of Michigan Press, 2002); Omar M.G. Keshk, Brian M.
Pollins, and Rafael Reuveny, “Trade Still Follows the Flag: The Primacy of Politics in a Simulta-
neous Model of Interdependence and Armed Conºict,” Journal of Politics, Vol. 66, No. 4 (novembre
2004), pag. 1155–1179, doi.org/10.1111/j.0022-3816.2004.00294.X; Valentin L. Krustev, “Interdepen-
dence and the Duration of Militarized Conºict,” Journal of Peace Research, Vol. 43, No. 3 (May 2006),
pag. 243–260, doi.org/10.1177/0022343306063930; Brett V. Benson and Emerson M.S. Niou, “Eco-
nomic Interdependence and Peace: A Game-Theoretic Analysis,” Journal of East Asian Studies,
Vol. 7, No. 1 (January–April 2007), pag. 35–59, doi.org/10.1017/S1598240800004847; Aysegul Aydin,
“The Deterrent Effects of Economic Integration,” Journal of Peace Research, Vol. 47, No. 5 (Septem-
ber 2010), pag. 523–533, doi.org/10.1177/0022343310370290; Håvard Hegre, John R. Oneal, E
Bruce Russett, “Trade Does Promote Peace: New Simultaneous Estimates of the Reciprocal Effects
of Trade and Conºict,” Journal of Peace Research, Vol. 47, No. 6 (novembre 2010), pag. 763–774,
doi.org/10.1177/0022343310385995; Solomon Polachek and Jun Xiang, “How Opportunity Costs
Decrease the Probability of War in an Incomplete Information Game,” International Organization,
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 15
lost as a result of war. The higher the opportunity cost, the less likely a state is
to resort to war. Generally, this opportunity cost is operationalized with the
full volume of trade between two states, which implies that all bilateral trade
is forfeited during war. Some research has focused on determining which por-
tion of lost trade actually constitutes a cost to the state. Mark Crescenzi models
economic interdependence based on the cost of switching to the next-best trad-
ing alternative; così, the opportunity cost is the “exit cost” from a market, non
the entire volume of bilateral trade.9 Han Dorussen disaggregates trade to the
industry level to assess in which industries loss of trade leads to higher oppor-
tunity costs.10 Cullen Goenner focuses on strategic resources as producing the
greatest opportunity costs, deªning “strategic resources” as energy, nonferrous
metals, chemicals, electronics, nuclear materials, and arms.11 While such stud-
ies reªne scholars’ understanding of which portion of trade is more important
for the state, they still assume that trade is severed during war, but that only a
part of this severed trade constitutes an opportunity cost.
Jack Levy and Katherine Barbieri hypothesize that the opportunity-cost
mechanism can work as political leverage in wartime.12 The state could
threaten to sever trade with the enemy in the midst of the war, knowing that
the enemy would have to pay a domestic cost for lost trade and therefore be
forced into suing for peace. This hypothesis is problematic. If severing trade
with the enemy constitutes enough leverage to coerce the opponent into suing
for peace, then it should also be enough leverage to prevent the opponent from
starting the war in the ªrst place.13 Because war is a costly enterprise, stati
would use existing leverage to avoid war, not wait to use the leverage during
war. Conversely, if the leverage is insufªcient to coerce the enemy into negotia-
tions before the war started, it is unlikely to do so during war, when the out-
come of the conºict is decided by armies on the battleªeld.
The second causal mechanism of economic interdependence focuses on pro-
Vol. 64, No. 1 (2010), pag. 133–144, doi.org/10.1017/S002081830999018X; and Copeland, Economic
Interdependence and War.
9. Mark J.C. Crescenzi, “Economic Exit, Interdependence, and Conºict,” Journal of Politics, Vol. 65,
No. 3 (agosto 2003), pag. 809–832, doi.org/10.1111/1468-2508.00213.
10. Han Dorussen, “Heterogeneous Trade Interests and Conºict: What You Trade Matters,” Jour-
nal of Conºict Resolution, Vol. 50, No. 1 (Febbraio 2006), pag. 87–107, doi.org/10.1177/0022002
705283013.
11. Cullen F. Goenner, “From Toys to Warships: Interdependence and the Effects of Disaggregated
Trade on Militarized Disputes,” Journal of Peace Research, Vol. 47, No. 5 (settembre 2010), pag. 547–
559, doi.org/10.1177/0022343310371881.
12. Levy and Barbieri, “Trading with the Enemy during Wartime,” pp. 13–14.
13. During the war, the enemy might be more desperate for the goods supplied by the enemy,
thereby increasing the leverage. D'altra parte, during the war, the enemy already could have
restructured its trade to deal with different partners, thus reducing any potential leverage.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 16
trade interest groups that seek to preserve their welfare gains by pressuring
public ofªcials not to take any action that would sever trade, including mili-
tary confrontation.14 This mechanism explains how the costs of cutting off
trade enter the war calculus, something the opportunity cost logic takes for
granted. The assumption that war severs trade still remains, Tuttavia.
The third causal mechanism deals with credible signaling, where ending
trade with an enemy sends a costly signal of resolve that a state is willing to
endure the high costs of war.15 Even here, if two states have started ªghting, Esso
is assumed that trade has already been severed as a signal to the other side.
Across these three causal mechanisms, economic interdependence makes the
absolute prediction that there should not be trade between belligerents in war.
The second body of literature that links economic gains to security concerns
focuses on relative gains.16 In this context, relative gains refers to how much
more beneªcial a trading relationship is to one trade partner compared to the
other. Because the gains from trade are convertible into military capabilities, UN
state with greater relative gains can ultimately develop greater military capa-
bilities than its trade partner. The more important relative gains are to a state,
the more likely the state is to cut trading ties. Scholars generally agree that
14. Polachek, “Conºict and Trade”; Paul A. Papayoanou, “Interdependence, Istituzioni, and the
Balance of Power: Britain, Germany, and World War I,” International Security, Vol. 20, No. 4 (Primavera
1996), pag. 42–76, doi.org/10.2307/2539042; Paul A. Papayoanou, Power Ties: Economic Interdepen-
dence, Balancing, and War (Ann Arbor: University of Michigan Press, 1999); Christopher Gelpi and
Joseph M. Grieco, “Economic Interdependence, the Democratic State, and the Liberal Peace,” in
Edward D. Mansªeld and Brian M. Pollins, eds., Economic Interdependence and International Conºict:
New Perspectives on an Enduring Debate (Ann Arbor: University of Michigan Press, 2003), pp 44–59;
and Etel Solingen, “Internationalization, Colaitions, and Regional Conºict and Cooperation,” in
Mansªeld and Pollins, Economic Interdependence and International Conºict, pag. 60–86.
15. James D. Morrow, “How Could Trade Affect Conºict?” Journal of Peace Research, Vol. 36, No. 4
(Luglio 1999), pag. 481–489, doi.org/10.1177/0022343399036004006; Erik Gartzke, Quan Li, E
Charles Boehmer, “Investing in the Peace: Economic Interdependence and International Conºict,"
International Organization, Vol. 55, No. 2 (April 2001), pag. 391–438, doi.org/10.1162/00208180151
140612; Arthur A. Stein, “Trade and Conºict: Uncertainty, Strategic Signaling, and Interstate Dis-
putes,” in Mansªeld and Pollins, Economic Interdependence and International Conºict, pag. 111–126;
and Nam Kyu Kim, “Testing Two Explanations of the Liberal Peace: The Opportunity Cost and
Signaling Arguments,” Journal of Conºict Resolution, Vol. 58, No. 5 (2013), pag. 894–919, doi.org/
10.1177/0022002713484280.
16. Joseph M. Grieco, “Anarchy and the Limits of Cooperation: A Realist Critique of the Newest
Liberal Institutionalism,” International Organization, Vol. 42, No. 3 (Estate 1988), pag. 485–507,
doi.org/10.1017/S0020818300027715; Michael Mastanduno, “Do Relative Gains Matter? America’s
Response to Japanese Industrial Policy,” International Security, Vol. 16, No. 1 (Estate 1991),
pag. 73–113, doi.org/10.2307/2539052; Duncan Snidal, “Relative Gains and the Pattern of Interna-
tional Cooperation,” American Political Science Review, Vol. 85, No. 3 (1991), pag. 701–726, doi.org/
10.2307/1963847; and Robert Powell, “Absolute and Relative Gains in International Relations The-
ory,” American Political Science Review, Vol. 85, No. 4 (1991), pag. 1303–1320, doi.org/10.2307/
1963947.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 17
wartime substantially increases the state’s focus on relative gains; Perciò, if
relative gains concerns were extrapolated to wartime, they would predict no
trade with the enemy in all wars.
Linking relative gains to trading with the enemy in peacetime, Peter
Liberman claims that states are more concerned with relative gains vis-à-vis
states that are near, powerful, offensively armed, and hostile.17 In war-
time, states will be highly sensitive to relative gains, because the opponent is
both armed and hostile, leading to the prediction that all trade with the enemy
should be severed. James Morrow takes the argument a step further to argue
that relative gains are inconsequential in peacetime, unless they are highly
unequal or can be converted into military advantage in secret.18 Whichever
state gains less from trade can use alternate sources of funding to maintain the
military balance of power. The same logic, Tuttavia, cannot apply to wartime,
especially to an unlimited war, as the majority of a state’s resources are dedi-
cated to the war effort, so even a small relative disadvantage can signiªcantly
inºuence the ultimate battleªeld outcome.19
Arguing that relative gains considerations affect military technology more
strongly than they do non-security goods, Levy and Barbieri hypothesize that
trade with the enemy in the latter category is more likely than in the former.
Aside from the empirical examples of weapons sold across enemy lines,20 there
are two reasons to question this explanation of wartime trade. Primo, it is almost
impossible to ªnd a clear deªnition of what counts as a military technology, es-
pecially in a world where dual-use technology is prevalent. Secondo, and com-
pounding the previous problem, trade in all products can ultimately be useful
for a state’s security. Any product traded to the enemy allows the enemy econ-
omy to free up domestic production from making those goods and instead fo-
cus resources on the war effort.21
Finalmente, the domestic political economy literature also suggests hypotheses
on trade with the enemy. At a very basic level, states receive absolute gains
from engaging in free trade with each other.22 Given purely economic motives,
trade with the enemy should continue—another absolute prediction. On the
17. Peter Liberman, “Trading with the Enemy: Security and Relative Economic Gains,” Interna-
tional Security, Vol. 21, No. 1 (Estate 1996), pag. 147–175, doi.org/10.2307/2539111.
18. James D. Morrow, “When Do ‘Relative Gains’ Impede Trade?” Journal of Conºict Resolution,
Vol. 41, No. 1 (1997), pag. 12–37, doi.org/10.1177/0022002797041001002.
19. Levy and Barbieri, “Trading with the Enemy during Wartime,” p. 10.
20. Ibid., pag. 10–12.
21. Schelling, International Economics.
22. David Ricardo, On the Principles of Political Economy and Taxation (London: John Murray, 1817).
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 18
other hand, the decision to open the economy to free trade creates winners and
losers in domestic politics.23 The beginning of a war could function as a focal
point, which domestic groups could use to change the state’s existing trading
policies to their preferred ones. If protectionist forces prevail in a state, trade
will be cut off during war. If free trade forces prevail in a state, trade will be al-
lowed to continue.
Additionally, severing trade with the enemy could do considerable damage
to a state’s domestic economy by creating food shortages, decreasing the reve-
nue from trade, increasing the costs of ªghting, or lowering foreign reserve
levels. This proposition coheres with the vulnerability argument made by
scholars arguing against the economic interdependence thesis.24 Kenneth
Waltz claims that depending on another state for the supply of vital resources
opens a state to threat if the supplier ever severs the trading relationship.25 For
Waltz, such potential dependency provides reason for great powers to pursue
policies of autarky. John Mearsheimer argues that such vulnerability explains
why states conquer others to ensure adequate supply.26 Regardless of the effect
on foreign policy, the relevant aspect for trade with the enemy is that states
might be compelled to continue trading during war to ensure an adequate
supply of necessary goods. I draw on similar ideas and explain how states de-
termine which products merit such protection during the war.
Levy and Barbieri propose that the nature of the war can affect the decision
to trade with the enemy; they are ambivalent, Tuttavia, on whether it increases
or decreases wartime trade.27 For example, they hypothesize that the longer
wars last, the more important relative gains from the early part of war be-
come, which provides incentives to sever trade. D'altra parte, the longer
the war lasts, the greater the cost of severing trade to the domestic economy
and private actors within the state, which incentivizes states to allow trade
with the enemy. Long wars also make restoring prewar trade ties harder, likely
leading to more trade with the enemy. Total wars have a greater impact on the
economy than do limited wars; così, out of relative gains considerations, Levy
and Barbieri propose that there should be less wartime trade during total
wars. My theory incorporates the length and intensity of war; Tuttavia, these
two variables alone cannot provide information about wartime trade. Both fac-
23. Wolfgang F. Stolper and Paul A. Samuelson, “Protection and Real Wages,” Review of Economic
Studi, Vol. 9, No. 1 (1941), pag. 58–73, doi.org/10.2307/2967638.
24. Copeland, Economic Interdependence and War.
25. Waltz, Theory of International Politics.
26. John J. Mearsheimer, The Tragedy of Great Power Politics (New York: W.W. Norton, 2001).
27. Levy and Barbieri, “Trading with the Enemy during Wartime,” pp. 28–30.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 19
tors work as thresholds against which states judge which products should be
allowed for trade with the enemy.
Overall, the hypotheses about wartime trade with the enemy drawn from
the existing literatures provide absolute predictions: either states should al-
ways trade with their enemies during war, or they should never trade. Some
individual variables taken from these literatures—the nature of the war, the se-
curity nature of the product, and domestic lobby groups—do allow for varia-
zione; Tuttavia, these variables lack a causal logic to explain when and how
they affect wartime trade. What is missing is a unifying theory that explains
which products states trade with their enemies, when during the war they
trade them, and for what reason.
Theory: Why States Trade with Their Enemies
A state allows trade with the enemy during wartime if either one of two condi-
tions is met: (1) when it does not help the enemy win the current war, O
(2) when ending trade would damage the state’s long-term security. Contro-
versely, trade is prohibited when both conditions are not met: trade aids the
enemy in the ongoing war, and it can be severed without damaging the state’s
future security.
To determine when a product meets these conditions, a state assesses the
characteristics of that product relative to the expected nature of the war. A
judge if a product helps the enemy in the course of the current war, a state
compares how long it takes the adversary to convert the item into military ca-
pabilities to the expected length of the conºict. Products that are convertible to
military capabilities only after the war ends can be safely traded with the en-
emy. Regarding the second condition, the expected intensity of the war largely
determines which revenue-generating products need to be traded with the en-
emy to ensure the long-term security of the state. The more intense the war, IL
more revenue a state will be willing to give up by cutting trade to guarantee its
survival in the current war. In this section, I start by delineating what is most
puzzling about wartime trade. Then I explain the product characteristics and
the war characteristics necessary for solving that puzzle, before turning to an
explanation of how these factors are combined to make a wartime commercial
policy. The last two sections address likely concerns with the theory.
the puzzle of wartime trade
Trading with the enemy during war can be dangerous. All trade carries secu-
rity externalities: the gains from trade can be converted by the opponent into
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 20
military capabilities.28 During war, the last thing a state wants to do is help
the enemy increase its warªghting capacity. Così, to increase the chances
of winning the current war, a state wants to sever all trade that carries secu-
rity externalities.
Allo stesso tempo, severing trade entails economic costs that affect the state
both during and after the war.29 First, the state loses access to the gains from
trade—the product being traded or the funds received from the trade. Secondo,
the state loses the income from taxing the circulation of the product within the
domestic economy. Inoltre, some economic costs persist after the war.
The state loses its market share in the enemy state if such trade can be replaced
by neutral states. Additionally, there is the relative loss vis-à-vis these neutral
stati, as they are now gaining from trade given up by the state.30 These costs
matter because they affect the ability of the state to fund its long-term security.
A disruption in the stream of revenue into state coffers stemming from sever-
ing trade—especially a permanent disruption—affects what the state can in-
vest in its own military capabilities.31 It would impede not only the ability of
the state to prosecute the current war, but also the state’s preparedness for fu-
ture wars. To ensure its ability to invest in long-term security, a state wants to
keep as much trade open as possible.
product-level characteristics
Wartime commercial policy seeks to reconcile these conºicting preferences—
minimizing security externalities by prohibiting trade that is counterproduc-
tive to the war effort, while maximizing revenue by allowing as much trade as
possible. To understand how states reconcile these incentives, two theoretical
insights are necessary.
Primo, security externalities are not instantaneous. It takes time for a state to
convert the gains from trade, whether from imports or exports, into military
capabilities. More important, this conversion time varies considerably by
Prodotto. The shortest practical conversion time is selling a gun to the enemy
across the battleªeld. Although incredibly rare, such situations have occurred
throughout history. At the beginning of the Siege of Grave in the Wars of
28. Gowa and Mansªeld, “Power Politics and International Trade”; and Gowa, Allies, Adversaries,
and International Trade.
29. Ricardo, On the Principles of Political Economy and Taxation.
30. Eugene Gholz and Daryl G. Press, “The Effects of Wars on Neutral Countries: Why It Doesn’t
Pay to Preserve the Peace,” Security Studies, Vol. 10, No. 4 (Giugno 2001), pag. 1–57, doi.org/10.1080/
09636410108429444.
31. Schelling, International Economics.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 21
Louis XIV, the French commander of Grave, at the instruction of the secretary
of state for war, sold gunpowder to the Dutch forces that were besieging
the town.32
A particularly long conversion time would stem from the import of a raw
material that is processed domestically for export. Per esempio, a state imports
raw gems, converts them into ªne jewelry, then sells the jewelry abroad. IL
conversion time includes the time it takes to make the deal for purchasing and
delivering the gems, the time to transport the gems to the factory, the time to
reªne and set them into the jewelry pieces, the time to transport the jewelry
to the buyer, and the time to receive payment for the sale. And at the end of
this process, the state would still have to convert the revenue from taxing this
domestic manufacturing process into military capabilities, which necessitates
additional time to purchase guns or food rations and deliver them to the bat-
tleªeld. Così, although trade in raw gems can ultimately aid the war effort,
converting them to military capabilities takes a long time.
The different products traded between two states can be ranked based
on their approximate conversion times into military capabilities. A product
closer to the end of a supply chain has a shorter conversion time than a prod-
uct at the beginning of the same supply chain. A ªnished good that is of imme-
diate use on the battleªeld (guns, ammunition, tanks, ªghter jets, eccetera.) should
have a faster conversion time than ªnished goods that are not. It is not neces-
sarily the case, Tuttavia, that products at the beginning of supply chains in
military production will have faster conversion times than products at the be-
ginning of supply chains in nonmilitary production. This depends on how
long it takes to get through the entire supply chain. It takes less than a day to
turn wheat into bread, whereas it takes about twenty-two months to go from
a bolt to an F-35.33 Even if the process of sowing grain and reaping the wheat
is considered, it still takes longer to manufacture an F-35 than to make a loaf
of bread.
Secondo, not all trade is equally important to the economy. The amount of
income lost to the state from severing trade, in either imports or exports,
32. The reason for this, no matter how counterintuitive, was to increase their chances of survival.
The quantity of gunpowder within their walls was too high, and just one accident would have
leveled the entire town. Dutch bombardment of the fortiªcation—even using the French
gunpowder—caused considerably less damage. Camille Rousset, Histoire de Louvois et de Son Ad-
ministration Politique et Militaire [History of Louvois and his political and military administration]
(Paris: Didier et cie, 1862), P. 65.
33. Evan Hoopfer, “Lockheed Martin Gears Up for Increased Production of F-35 in Fort Worth,"
Dallas Business Journal, Giugno 7, 2017, https://www.bizjournals.com/dallas/news/2017/06/07/
lockheed-martin-f-35-fort-worth.html.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 22
also varies by product. Severing trade in products with close substitutes or
sources of supply other than the enemy, regardless of how important these
products are to the economy, would not have a large impact on the con-
tinuity of funds available to the state.34 There is a one-time switching cost re-
quired to adjust the industry to working with the substitute product; IL
industry will continue to function and produce revenue, Tuttavia. Compared
to severing trade in the product outright, here the income of the state is only
slightly affected.
If a product is neither substitutable nor has alternative sources of trade, pro-
hibiting trade will mean a loss of all tax revenue generated through the circula-
tion of this product in the economy. The amount of revenue lost can be ranked
based on the number of points at which a speciªc product can be taxed—
import tax, corporation tax if the item is involved in a domestic production
chain, export tax, personal tax if it is imported and sold by individual
merchants, sales tax if sold domestically after import, and so on. The relative
rates of taxation at each of these points inºuence which products contribute
more to the revenue of the state; in general, Tuttavia, the more taxation points,
the more revenue a product can bring to a state. The import of a motorcycle
(ªnished good) should bring in less revenue than the import of steel (raw ma-
terial) involved in the domestic car industry. Whereas the motorcycle can be
taxed only at the moment of importation, the steel also can be taxed through
each step of the manufacturing process and when it is ultimately sold as a
ªnished car. Allo stesso modo, the export of raw materials should bring in less revenue
to a state than the export of ªnished goods. If the export of a product that is
manufactured domestically but for which there is no domestic demand is pro-
hibited, in the long term, the production of this product would stop, depriving
the state of revenue from this industry. Overall, products that require domestic
production contribute more to state revenue than products that do not (Vedere
ªgure 1).
These two theoretical insights help reconcile the two incentives facing the
state. Some products take a long time to be converted into military capabilities.
Generalmente, these products are more likely to be traded with an enemy during
war. The security concern is to prevent the enemy from increasing its capabili-
ties for the current conºict; if the opponent never has a chance to do that, trade
can continue. Some products can be easily substituted or diversiªed away
34. Katja B. Kleinberg, Gregory Robinson, and Stewart L. French, “Trade Concentration and Inter-
state Conºict,” Journal of Politics, Vol. 74, No. 2 (April 2012), pag. 529–540, doi.org/10.1017/s00223
81611001745; and Goenner, “From Toys to Warships.”
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 23
Figura 1. Distribution of Traded Products Based on Enemy’s Conversion Time and a
State’s Loss of Revenue
less likely to be prohibited
= traded product
steel
more likely to
be prohibited
e
tu
N
e
v
e
R
gun
time
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
from the enemy. If security concerns dictate that trade in these products is dan-
gerous in the current conºict, severing this trade would not create too much
disturbance in the state’s long-term security. Per esempio, in considering trade
with Japan during the Russo-Japanese War (1904–05), Russia was very likely to
have prohibited trade in guns with the enemy. Japan could convert guns
to military capabilities very quickly—it simply had to transport them to the
battleªeld, and the export of guns was not a very important industry for
the Russian economy. Steel, on the other hand, was more likely to have
been traded with the enemy. It would have taken longer to process steel into
military capabilities, and Russia was a steel-exporting country at the time.35
war-level characteristics
Although product characteristics can be used to rank products based on
their likelihood of being traded with an enemy, they are not sufªcient to ex-
plain a state’s wartime commercial policy. Both of the factors that determine
a wartime commercial policy—whether the trade helps the enemy win and
35. Malcolm R. Hill, “Russian Iron Production from the Repeal of Serfdom to the First World
War,” Icon, Vol. 22 (2016), pag. 115–138, https://www.jstor.org/stable/44242743.
International Security 46:1 24
whether the state can withstand the loss of the trade—depend on the condi-
tions of the war being fought. Just as states choose different military strategies
for different kinds of wars, states choose different wartime commercial poli-
cies to match the speciªc conºict they are expecting to ªght. The two dimen-
sions of war relevant to trade with the enemy are the expected length and the
expected intensity.36
War conditions allow the state to make threshold decisions about which
products should be traded. Per esempio, what if a product contributes little to
the income of the state, but has a very long conversion time? The state can
withstand the loss of trade, but this is not likely to lead to military beneªts.
This decision is made with respect to the expected length of the war. What if a
product contributes greatly to state revenue but has a short conversion time?
Continuing trade will assist the enemy, but severing trade will damage the
state’s long-term security. This decision is made with respect to the expected
intensity of the war.
The expected length of the war is the decisionmakers’ estimate about how
long the current war will last—is the state expected to win a quick victory or to
face a prolonged struggle? This estimate can be based on the balance of capa-
bilities between major belligerents,37 the development of new war-winning
technologies or strategies,38 analogy or learning from previous conºicts, E
so forth. Empirically, states frequently fall prey to the “allure of battle”
and assume they will win a quick victory, even though most wars fought in
human history have been won by attrition.39 States preparing for war tend to
start with the expectation of a short struggle ended by a decisive battleªeld
victory.40 As the war reveals more information, states update their expecta-
tions about its expected length. Optimism for the success of a new offensive
can decrease the expected duration of a war; failure of the offensive or a suc-
cessful counteroffensive by the enemy can increase it.
To decide if a speciªc product can help the enemy change battleªeld out-
comes, a state compares the expected length of the war to how quickly the op-
36. This is similar to the typology of war used by Alex Weisiger; Tuttavia, his work classiªes wars
after the fact, whereas this project focuses on decisionmakers’ assumptions about a war before it
happens. Weisiger, Logics of War: Explanations for Limited and Unlimited Conºicts (Ithaca, N.Y.:
Cornell University Press, 2013).
37. Ivan Arreguín-Toft, “How the Weak Win Wars: A Theory of Asymmetric Conºict,” Interna-
tional Security, Vol. 26, No. 1 (Estate 2001), pag. 93–128, doi.org/10.1162/016228801753212868.
38. John J. Mearsheimer, Conventional Deterrence (Ithaca, N.Y.: Cornell University Press, 1985).
39. Cathal J. Nolan, The Allure of Battle: A History of How Wars Have Been Won and Lost (Oxford: Ox-
ford University Press, 2017).
40. Jack Snyder, The Ideology of the Offensive: Military Decision Making and the Disasters of 1914
(Ithaca, N.Y.: Cornell University Press, 1984).
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 25
ponent can convert this product into military capabilities. Products with
conversion times longer than the expected length of a war are allowed to be
traded. Simply put, if the state plans to win before the opponent can beneªt
militarily from trade, there is no reason to sever trade. As the war progresses,
a state updates its expectations about the length of the war, which leads to a
recalibration of its wartime commercial policy. Each time a state is forced to ex-
tend its assessment of the length of the war, more products should be prohib-
ited from trade. The newly prohibited products would have conversion times
shorter than the new expectation of the length of the war. Conversely, when a
state once again expects to quickly win the war, it may remove some products
from the prohibited trade lists.
The expected intensity of war refers to how existentially threatening the
decisionmakers expect the war to be. Is state death a possible outcome? Fa
the war involve the invasion of a state’s homeland or a ªght in a distant re-
gion? Is the war over a vital national security interest or a matter of less stra-
tegic importance?41 The more existentially threatening war becomes, IL
more the state will sacriªce its future security to ensure its immediate survival.
Translated into economic terms, the more intense the war, the more the state
will be willing to lose income from key industries—which decreases invest-
ment in its future security—to prevent the opponent from beneªting militarily
from trade—which increases the state’s chances of winning the war. The inten-
sity of a war can change during the course of the conºict. Factors that mark a
war’s increasing intensity include the ªghting shifting closer to the state’s cap-
ital, increasing fears of decisive defeat on the battleªeld, and the opponent
adopting maximalist war aims or exhibiting ideological fanaticism. All these
factors would make a state more willing to sacriªce its future security to guar-
antee its survival in the present war. D'altra parte, when additional allies
join the ªght, each member of the expanding alliance might think that the in-
tensity of the war has lessened, making them less willing to sacriªce various
aspects of their trade.
The intensity of the war helps a state navigate the trade-off between dedicat-
ing all its effort to winning the current war at the cost of undermining its fu-
ture security and dedicating all its efforts to long-term security at the cost of
potentially losing the current war. To judge whether it can afford to sever trade
41. Robert Endicott Osgood, Limited War: The Challenge to American Strategy (Chicago: Università
della Chicago Press, 1970); Nordal Åkerman, On the Doctrine of Limited War (Lund, Sweden:
Berlingska Boktryckeriet, 1972); and Stephen Peter Rosen, “Vietnam and the American Theory of
Limited War,” International Security, Vol. 7, No. 2 (Autunno 1982), pag. 83–113, doi.org/10.2307/2538434.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 26
in a speciªc product in an ongoing war, the state compares the expected inten-
sity of the war—that is, the level of investment the state deems necessary to
ensure its long-term security—to how much income the product generates by
circulating in the economy. Products that generate more income than the state
is willing to sacriªce are allowed to be traded. Inoltre, to ensure its long-
term security, the state will continue trading such products, even if the enemy
can convert them into military capabilities that can be employed in the present
conºict. Ancora, as the expected intensity of war increases and the very survival
of the state is threatened, the state will bear greater losses of income to win. In
such cases, the state will cease trading even products that are essential to key
industries, prioritizing immediate survival over long-term security.42
Per esempio, in a state with an economy dependent on the textile industry,
the import of raw cotton would contribute greatly to the state’s revenue. In a
limited war with the supplier of this cotton, the state would prefer to continue
importing the cotton. The income generated from the textile industry is neces-
sary for the economy and thus for the continued investment in military capa-
bilities. If the war becomes existentially threatening, Tuttavia, the state would
be willing to sever trade in cotton—thus damaging its long-term security—to
deprive the enemy of funds it receives from trade in cotton—thus damaging
the opponent’s ability to wage the current war and increasing the state’s
chances of survival.
the formation of a wartime commercial policy
Figura 2 brings together the two factors affecting a state’s wartime commercial
policy. The axes of the chart denote product characteristics. The plot points
represent individual products. When a state is formulating a wartime commer-
cial policy, the decision to continue trading is made separately for each prod-
uct. Each product that is traded between enemies before the war starts can
be mapped, based on how fast the enemy can convert the gains from trade
into military capabilities and how important the product is to the state’s do-
mestic economy.
The thresholds on the chart are characteristics of the war. The same state
in the same year can create different wartime commercial policies, depending
on how long and how intense it expects a speciªc war to be. Based on these ex-
pectations about the war, all products that fall within the bottom left quadrant
(IO) should be prohibited from trade during war. These are the products in
42. Survival is the primary goal of the state insomuch as without it, no other goal is achievable.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 27
Figura 2. A State’s Wartime Commercial Policy (TWE (cid:2) trade with the enemy)
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
which the state can afford to lose trade and that would contribute to the oppo-
nent’s military capabilities in the current conºict.
The state can afford to lose trade in products in the bottom right quad-
rant (III), but does not have any military incentives to sever the trade. Come il
opponent will not have time to convert these products into military capabili-
ties, they can be safely traded. Trade in products in the top left quadrant (II)
is too important for the state to lose. Even though the opponent beneªts mili-
tarily from the trade, the state has to maintain trade in these products to ensure
its long-term security. Products in the top right quadrant (IV) should be al-
lowed for trade for both reasons.
Additionally, ªgure 2 facilitates visualization of changes to a state’s wartime
commercial policy. If the expected length of the war increases, the vertical
threshold moves to the right. Based on this, the state would increase the num-
ber of products that are prohibited from trade. Some of those products that
were previously in the bottom right quadrant (trade allowed) would become
part of the bottom left quadrant (trade prohibited). Allo stesso modo, if the intensity of
war increases, the horizontal threshold moves up. This would also lead the
state to increase the number of products that are prohibited from trade. Some
products that were previously in the top left quadrant (trade allowed) would
become part of the bottom left quadrant (trade prohibited).
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 28
is it not endogenous?
Some observers might raise concerns over endogeneity, as the prohibitions on
trade could plausibly affect the expected length of a war. This is not a problem
for this theory for two reasons.
Primo, a prohibition on trade with the enemy is a long-term policy instru-
ment. The effects of the policy are felt only after it has been enforced for some
substantial period of time. After the effects are felt by the enemy, the economic
pressure increases proportionally with the length of time the policy is en-
forced. In the short term, Tuttavia, a prohibition on trade does not have an ef-
fect. The enemy can have strategic stockpiles of prohibited products, E
thereby not feel the effects of severed trade. Likewise, states have alternative
sources of income, so a lack of proªt from prohibited trade can, in the short
term, be made up with alternative methods of war funding. It is only once the
initial stockpiles of products (or proªts) are exhausted and the supply of addi-
tional trade is severely reduced that the economic pain of trade sanctions
can be felt.43 Thus, an initial expectation of a short war cannot stem from the
chosen wartime commercial policy.
Secondo, if the expectation of the length of the war and the prohibitions on
trade inºuence each other, no decisions on wartime trade could ever be
reached. Consider one traded product that is on the threshold of a war-length
expectation. If there were an endogeneity problem, prohibiting trade in this
product would shorten the expected length of the war. Lowering the expected
length of the war, Tuttavia, also means that the product in question should
now be allowed to be traded with the enemy. If it is traded with the enemy, IL
war-length expectation rises again, and so the product should not be traded
with the enemy. This cycle continues indeªnitely. The very fact that decisions
on wartime trade are made shows that the judgments about the length of the
war and the prohibitions on products are made separately.
overlap between belligerent wartime commercial policies
My theory explains how a state forms a wartime commercial policy toward an
enemy belligerent. A state weighs the beneªt of severing trade in products that
43. It took almost four years to starve Germany in World War I—from March 1915 through 1919.
Alexander B. Downes, Targeting Civilians in War (Ithaca, N.Y.: Cornell University Press, 2011),
pag. 83–114. Likewise, it required nearly four years to starve Japan in World War II. Sheldon Garon,
“The Home Front and Food Insecurity in Wartime Japan: A Transnational Perspective,” in
Hartmut Berghoff, Jan Logemann, and Felix Römer, eds., The Consumer on the Home Front: Secondo
World War Civilian Consumption in Comparative Perspective (Oxford: Oxford University Press, 2017),
pag. 29–54; and Christopher Clary, “The Starvation Myth: The U.S. Blockade of Japan in World
War II,” Fairmount Folio: Journal of History, Vol. 4 (2000), pag. 125–137, https://journals.wichita.edu/
index.php/ff/article/view/62/69.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 29
can be converted by the opponent into military capabilities against the cost of
losing the income from this trade. In definitiva, it divides all products into those
allowed to be traded with the enemy and those that are prohibited. Al
same time, the opponent is making a similar calculation to determine its pre-
ferred wartime commercial policy. Because trade requires at least two willing
participants, what trade will actually exist between two belligerents during
war depends on where their preferences overlap.
At ªrst glance, it might seem that the commercial policies of two belligerents
should be polar opposites—every product that one state would want to con-
tinue to trade in, the opponent would have reason to sever. My theory pro-
vides ample opportunity, Tuttavia, for two states to retain the same portion of
their trade.
Primo, there are two different reasons why a state might trade with the en-
emy. State preferences for allowed trade can overlap, because the two states’
reasons for trading are different. State A might continue importing product X
because it is a raw material necessary for a key domestic industry—it is too im-
portant to the state economy to prohibit trade. State B might continue export-
ing product X because it does not expect State A to beneªt militarily from the
trade in the current war—the trade does not affect battleªeld outcomes.
Secondo, both states might continue trading with each other in a certain prod-
uct for the same reason. Both belligerents can be convinced that the war will be
short. They can, Perciò, allow considerable portions of trade to continue as
the security externalities would not affect the battleªeld outcome. Allo stesso modo,
both belligerents can consider their opponent to be inefªcient in converting the
gains from trade into military capabilities, allowing much trade to continue.
One or both of the states are likely mistaken in their assessment; nevertheless,
trade can continue.
Finalmente, both belligerents could ªnd trade in the same product indispensible.
Per esempio, during the Crimean War, Britain’s economy depended, among
other things, on importing ºax, hemp, linseed, and tallow, for which Britain
could not ªnd an alternative supply other than Russia.44 Russia could not ªnd
an alternative market to sell these goods, as Britain absorbed nearly half of
Russia’s European exports.45 Both states were unwilling to sever trade in these
prodotti, even as the war dragged on.
44. Olive Anderson, “Economic Warfare in the Crimean War,” Economic History Review, Vol. 14,
No. 1 (Gennaio 1961), pag. 34–47, doi.org/10.2307/2591352.
45. Olive Anderson, A Liberal State at War: English Politics and Economics during the Crimean War
(New York: Macmillan, 1967); and J.L. Ricardo, The War Policy of Commerce (London: Efªngham
Wilson, 1855).
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 30
Research Design
I test the theory with archival evidence from World War I, using the formation
and changes in British wartime commercial policy throughout the conºict.
World War I is a crucial case based on a least likely case design. The conven-
tional wisdom on trade with the enemy states that when concerns over the se-
curity of the state are high, trade with the enemy is least likely to occur. World
War I was a global, total war where the full might of each nation was used in
the war effort; it was a long war fought for the survival of the state. Showing
that trade with the enemy can occur even under such unlikely conditions lends
support to the argument that wartime trade is quite prevalent.
Focusing on Britain allows me to take advantage of the temporal variation in
how decisionmakers reassessed the expected length and intensity of the war.
The availability of uniquely granular, product-level data in the British case al-
lows me to test the product-level predictions of the theory as well as the war-
level predictions about a state’s wartime commercial policy.
To test the theory, I examine Britain’s wartime commercial policy at two
stages. Primo, the state grappled with the question of whether trade with the en-
emy should be allowed during a hypothetical future war and then adapted
this general strategy to the expected war with Germany. By process tracing
committee discussions, I identify Britain’s reasons for and against trade
with the enemy, noting which arguments prevailed. Additionally, after the
general policy was formed, I examine how it developed into a speciªc policy
toward Germany for the coming war. Secondo, I look at how this policy was
amended during the war. As the theory predicts that a state should change its
wartime commercial policy in response to battleªeld realities, I search for peri-
ods when decisionmakers reassessed their wartime commercial policy and
what changes they implemented. Speciªcally, was the policy becoming more
lax or more restrictive, and were licenses granted to legally circumvent the ex-
isting prohibitions?
As the discussions in committee about the contents of a wartime commercial
policy are necessarily general, they cannot help test the theory at the product
level. The goal at this level of decisionmaking is to set a guiding policy, as op-
posed to determining the fate of each traded product.46 To test the product-
46. A new bureaucracy is ultimately set up to determine whether each product should be traded
during the war. Unfortunately, the written record from these committees was not deemed impor-
tant enough to be preserved, with only a few “stereotypical” cases left in the historical record for
posterity.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 31
level predictions of the theory, I collected the ofªcial proclamations, publicized
to domestic merchants through the London Gazette, a government journal
that details the products prohibited from trade during war. The full collection
of these proclamations throughout the war forms Britain’s implemented war-
time commercial policy, allowing me to see how the policy changed at the
product level.
These data have some limitations. Primo, I do not have access to a list of
all products that were traded by British merchants. Trade records and tariff
documents have only aggregated categories, not speciªc products. The dataset
includes all products that were at some point prohibited from trade. For prod-
ucts prohibited later in the war, I know they were allowed to be traded before
that prohibition. If a product was never speciªcally prohibited from trade, IO
would be unable to account for it. In essence, I would be undercounting the ex-
tent to which trade with the enemy was permitted. Additionally, some level of
aggregation still exists in the data. Sometimes prohibitions occur by category.
Per esempio, “accessories for use in connection with aircraft” is a line item
prohibited from trade, but it refers to a category of traded goods. Through
other prohibitions, it is possible to tell that “barographs” and “revolution indi-
cators” are types of such accessories, but it is impossible to determine the other
products that would also fall into this category. Whenever possible, categorie
are disaggregated to the lowest possible level.
Finalmente, the proclamations list both prohibitions on exports and prohibitions
on imports of products. It is only possible, Tuttavia, to disaggregate which
products were speciªcally prohibited from trade with the enemy on the export
side. Britain prohibited products with three types of geographic scopes: cannot
be exported from Britain (to prevent shortages in the British war effort), can be
exported only to the colonies, and cannot be exported to countries adjacent to
the enemy (to prevent products from reaching the enemy). The geographic
scope of the prohibitions on some products changed throughout the war. Al-
though there could be multiple reasons to ban products from leaving Britain,
I can be most conªdent that the reason for prohibiting export to countries adja-
cent to the enemy was to sever enemy trade. Similar geographic distinctions
do not exist for prohibitions on imports, which were always prohibited from
all sources. As such, it is impossible to distinguish imports prohibited to stop
trade with the enemy and imports banned for other reasons. To keep the anal-
ysis cleanly focused on trade with the enemy, the product-level predictions of
the theory are tested only with export trade. Nevertheless, the case analysis
presents some product-level information on imports from the enemy focused
on licenses granted to trade during the war.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 32
The conversion time of each product is operationalized by its relative posi-
tion in the manufacturing process: ªnished goods, intermediate goods, raw
materiali, or substitute goods. Substitute goods are those used in manufactur-
ing after the goods they are replacing are no longer available. Products at the
beginning of a supply chain have longer conversion times than products at
the end of the same supply chain. Substitute goods should have the longest
conversion times relative to other products in the supply chain. It is impossible
to get a temporal measure of products’ conversion times, as this would require
industry knowledge of production processes of World War I–era technology as
well as the transportation routes for individual companies’ supply chains. IL
British government itself received only approximate estimates of such infor-
mation from industry representatives.47 An individual product’s contribution
to state revenue is likewise an elusive target. As such, testing this aspect of the
theory at the product level is left for future research.
The expected length of the war is based on how long the state expected to
keep troops on the battleªeld. If decisionmakers anticipated that ªghting
would take less than one campaign season without troops having to winter
near the battleªeld, the war is considered short. Alternatively, if they expected
ªghting to require more than one campaign season, the war is considered long.
To code this expectation, I gathered month-by-month information from two
fonti. The ªrst is the expectations of British decisionmakers about the length
of the war, collected from secondary literature. What sort of war were these
leaders preparing to ªght? How long of a campaign were they expecting?
How optimistic were they about achieving their goals in the set time frame?
Did they think about the necessity of calling up reservists? The second source
of information is the actual arrangements made for the war effort: Were costly
investments made in the war that might suggest preparations for a long con-
ºict? Per esempio, the construction of extensive fortiªcations, the preparation
of trenches, and the construction of ªeld railroads to solidify supply lines
would indicate a state preparing for a long war. All these actions would be a
waste of resources and effort if the war was expected to be short. On the other
hand, sending troops to a location with a cold climate without winter clothes
would suggest a state preparing for a short struggle.
47. The lack of timely information experienced during World War I was used to justify a request
for more personnel with industry expertise for the Ministry of Economic Warfare. “C.R.T. War
Work,” April 6, 1938, BT 11/966, British National Archives (BNA), pag. 21–28.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 33
British Wartime Commercial Policy in World War I
Britain’s initial wartime commercial policy in World War I carefully balanced
economic and military concerns about trade. Whereas direct trade with the en-
emy was prohibited for practical and nationalistic reasons, indirect trade was
almost entirely permitted. Trade in some products—speciªcally coal tar dyes,
sugar, and metal manufactures—was identiªed as being too important to the
British economy to sever. These goods were key to keeping domestic manufac-
turing running and, as such, contributed greatly to the income of the state.
Other products were deemed “innocent” goods used exclusively in civilian
manufacturing. Because Germany, Britain’s primary expected enemy, would
be unable to beneªt from trade in “innocent” goods militarily, Britain could
continue to supply them. D'altra parte, products including guns, ammu-
nition, gunpowder, airplanes, and surgical bandages were deemed too dan-
gerous to trade with the enemy, as they could be converted into military
capabilities quickly. Not all military equipment, Tuttavia, was on the initial list
of prohibited trade with the enemy, as trade in machine guns, shipbuilding
materiali, and implements for digging trenches was initially permitted.
Britain made this policy with a quick war in mind and signiªcantly
amended it when this assessment proved faulty. The longer the war was ex-
pected to last, the more products were added to the list of prohibited trade.
The more existentially threatening the war became, the more Britain was
willing to give up trade in products important to its own economy. Around
February/March 1915, after British leaders realized that ªghting on the
Western Front was stalled and that an injection of new men and new supplies
would not break this stalemate, food, raw materials, and a number of items
that were previously deemed “innocent” were added to the prohibition list.
By the end of 1916, with the end of war nowhere in sight and the hopes of vic-
tory starting to dwindle, Britain severed trade in the majority of the products
that it previously had deemed were too important to the economy.
assessment of the commercial situation
In 1911, in preparation for a potential war with Germany, the British govern-
ment created the Standing Subcommittee of the Committee of Imperial
Defence on Trading with the Enemy (hereafter Committee on Trading with
the Enemy). Its purpose was to examine “the subject of trade between
British subjects and the enemy in time of war, the extent to which it should be
allowed and the means and method of carrying out the policy recommenda-
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 34
tions.”48 After developing the general guidelines for wartime trade, IL
Committee was instructed to create the wartime commercial policy that Britain
would use in the looming war with Germany. One of the necessary pieces of
information for this examination was the importance of trade between Britain
and Germany. The Board of Trade provided such an assessment.49 In 1910,
German goods imported into and retained in the UK were valued at £58 mil-
lion, slightly less than one-tenth of
imports into the United
Kingdom.50 At the same time, Britain exported £37 million of its produce
and manufacturing to Germany, which amounted to about one-twelfth of its
total exports.51
the total
Of the goods that Britain imported from Germany, around 44 percent were
products for which Germany was the predominant supplier.52 These included
unreªned beetroot sugar, coal tar dyes, cotton piece goods, cotton hosiery,
mixed silk goods, furs, toys and games, iron and steel manufactures, steel
sheet bars, and tinplate bars.53 Three of these categories—sugar, coal tar dyes,
and metal manufactures—were important raw materials in British manufac-
turing. The rest were primarily ªnished goods that required little processing in
the UK. While the Board of Trade believed that domestic manufacturing could
have ªlled the domestic demand for clothing, rubber, and earthenware prod-
ucts, it would have been insufªcient to ªll the demand for clocks, coal-tar
dyes, electrical glow lamps, mixed silk goods, and sugar.54 In the case of sugar,
the situation was most dire, as the Board of Trade believed that if Germany
was effectively prevented from exporting sugar, there would not be enough
global supply left to cover British needs.
The Board of Trade assessed trade in products ºowing from Germany to
Britain as being mutually dependent—Germany needed Britain as a market,
and Britain needed Germany as a source of goods. Trade in the opposite
48. “Terms of Reference," 1912, Report and Proceedings of the Standing Subcommittee of the Committee
of Imperial Defense on Trading with the Enemy, CAB 16/18A, BNA, P. vi. (Hereafter, this report is re-
ferred to as Report.)
49. This case study relies solely on information gathered by the Board of Trade for the Committee
of Trading with the Enemy, as this was the information that the committee used to make its policy
recommendations. The accuracy of either the statistical information or the conclusions drawn by
the Board of Trade is largely irrelevant, as it was the only expert information provided to the
committee.
50. “Memorandum by the Board of Trade on the Probable Effects of a War with Germany on Brit-
ish Trade, No. 8,” December 1911, Standing Sub-Committee of the Committee of Imperial Defence
(SSCID), WO 106/45, BNA, P. 2.
51. Ibid.
52. “Predominant” refers to more than 50 per cento. Ibid., P. 13.
53. Ibid.
54. Ibid.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 35
direction, from Britain to Germany, Tuttavia, was assessed differently. For
Germany, imports from the UK accounted for around 8.5 percent of its total
imports; for nearly two-thirds of the products it received from the UK, how-
ever, the UK was the predominant supplier.55 These items included salted her-
ring, yarn of wool and hair, heavy woolen tissue, cotton yarn, cotton tulle,
cotton tissues, sheep and lambskins, tinplate, and cotton-spinning and prepar-
ing machinery.56 Although this trade would have been difªcult for Germany to
replace, severing it would not have applied maximum pressure on Germany,
as none of these products were raw materials indispensable to industry or es-
sential food items.
It was also assumed, by the Board of Trade, that Germany was not entirely
necessary to Britain as a market for British goods. Less than one-sixth of the
value of exports from Britain to Germany consisted of products for which
Germany was the predominant market.57 These items were maize meal, sharps
and middlings, fresh and cured herring, painters’ colors (barites), thrown silk,
slates for rooªng, wool ºocks, worsted yarns, alpaca and mohair yarns, E
yarn of wool or hair.58 Although speciªc groups of merchants in Britain would
have been adversely affected by the loss of this trade, overall, Britain
would not have suffered much.
The Board of Trade conducted a more thorough investigation of the wool
(main German import from Britain) and sugar trades (main British import
from Germany). It was found that, in addition to the wool imported into
Germany from the UK, most German wool came from British colonies. Al-
though South American wool could possibly have been used as a substitute, Esso
would have required considerable upfront costs, as Germany did not yet have
the machinery necessary to process that type of wool.59 A thorough assessment
of the sugar trade led to the conclusion that only Britain was dependent on the
import of sugar. The position of sugar manufacturing in Germany was of mi-
nor importance, and Germany could easily use the sugar it produced to satisfy
domestic demand.60 Thus, Germany required British wool, and Britain re-
quired German sugar.
In making its assessments, the Board of Trade fully recognized that neutral
55. Ibid., P. 9.
56. Ibid., P. 10.
57. Ibid., P. 11.
58. Ibid.
59. “The Wool Trade and Germany”, Febbraio 7, 1912, Report, P. 394.
60. “The German Sugar Trade and the Effect of Prohibiting the Importation of German Sugar,"
Febbraio 15, 1912, Report, P. 415.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 36
states would pick up the trade that would be forbidden to British and German
ships. If trade with the enemy was allowed, it was assumed that trade would
continue between the two countries directly using neutral ships; if it was pro-
hibited, trade would continue indirectly through neutral ports.61
assessment of the length of the war
Another signiªcant piece of information that the Committee on Trading with
the Enemy wanted to have, to make appropriate policy recommendations,
was the probable duration of war between the Triple Entente and the Triple
Alliance. A note from March 12, 1912, from the General Staff, while refusing to
go into any particular details on the subject, stated that “it would not be safe
to calculate on the war lasting less than six months.”62 The War Ofªce did not
want to give a point prediction for the end of the war to avoid being blamed
later for a mistaken prognosis. Additionally, the expected length of the war
inºuenced the extent of military stockpiles to be kept in the country. Preparing
for a shorter war would have assured the military access to fewer resources
than a longer war. Given this incentive structure for the War Ofªce, IL
Committee on Trading with the Enemy assumed that the conºict would not be
especially prolonged.
setting the initial wartime commercial policy
The members of the Committee on Trading with the Enemy, who came from
various branches of government, interpreted the information presented by the
experts differently and reached different conclusions about the optimal British
wartime commercial policy. There were two signiªcant points of agreement,
Tuttavia. Primo, the selected policy had to match the type of war. Secondo, differ-
ent types of products had to be treated differently.
All members of the Committee on Trading with the Enemy understood that
economic pressure was a time-dependent policy instrument. Lord Desart, IL
chairman of the Committee, pointed out that “a short period of extreme com-
mercial pressure must affect the permanent commercial interests of the nation
less than a prolonged period during which that pressure, which must in any
case be heavy, was reduced to its lowest possible proportion.”63 Even if the
Committee had resolved to sever all trade with the enemy and picked the most
stringent method of enforcing this policy, in the short term, it would have
no effect on helping Britain win the war. Only if the war were to become pro-
61. “Memorandum by the Board of Trade,” December 1911, pag. 3–8.
62. “Note by the General Staff,” March 12, 1912, Report, P. 424.
63. “Note by the Chairman,” February 21, 1912, Report, P. 419.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 37
longed would commercial policy be able to contribute to battleªeld outcomes.
Sir R. Chalmers, a representative from the Treasury, emphasized that “if
the war were brought to a rapid conclusion by the success of the German
armies over France,
it would not be worthwhile for us to exercise the
weapon of economic pressure, which must necessarily require time to produce
effect.”64 Similarly, if allies were to defeat Germany quickly, most of the eco-
nomic pressure that came from regulating trade with the enemy would not
even begin to be felt.
Allo stesso tempo, the Committee on Trading with the Enemy was planning
for a short struggle. Infatti, the secretary of the Committee, summarizing the
scope and ground to be covered, stated: “It is of course utterly impossible to
foresee what developments of the policy to be adopted may become necessary
during the course of a protracted war. Probably all that can be done now is to
consider the most suitable policy for adoption at the outset of a war.”65 The
Committee recognized that the policy it created had to match the type of war
Britain expected to ªght. And if the type of war changed, the policy had to
change with it.
In addition to agreeing that the wartime commercial policy must suit the
speciªc war, the members of the Committee on Trading with the Enemy de-
cided that different types of products should be treated differently by that
policy. They divided trade into three conceptual categories: arms, food and
raw materials, and “innocent” goods. Regardless of how open or restricted a
member of the Committee wanted the ultimate wartime commercial policy to
be, everyone agreed that arms, ammunition, and military and naval stores
would be prohibited from trade with the enemy.66 The opponent could
quickly increase its military capabilities with such products even in a fairly
short conºict. “Innocent” goods, likewise, did not pose any difªculty for
policymakers. As noted earlier, goods classiªed as “innocent” were those
used exclusively in civilian manufacturing.67 They were assumed not to as-
sist the enemy in prosecuting the war, and as such could be safely traded with
the enemy.68
Raw materials were the contentious category. While it was agreed that these
products could not beneªt the enemy militarily in a short period of time, Essi
64. “Minutes of Sixth Meeting,” February 23, 1912, Report, P. 86.
65. “Note by the Secretary, No. 1,” February 6, 1911, SSCID, WO 106/45, BNA, P. 1.
66. Not everyone agreed on which products ªt this description, and several meetings were held to
discuss precisely this question. In definitiva, the products were a compilation of ªnished goods that
were of immediate use on the battleªeld.
67. “Draft Report,” February 21, 1915, Interdepartmental Committee on Trading with the Enemy, CAB
42/1/46, BNA, P. 1.
68. “Report,” July 30, 1912, SSCID, CAB 38/21/31, BNA, P. 5.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 38
were assumed to increase the revenue of the enemy substantially.69 The debate
was over whether the enemy’s revenue from trade should be a relevant factor
in making wartime commercial decisions. In its ªnal report, the Committee on
Trading with the Enemy settled on recommending a policy of prohibiting raw
materials from trade.70 This part of its recommendation was not accepted by
the Cabinet, Tuttavia. Although prohibiting the enemy from gaining raw ma-
terials could exert great pressure, as the Committee on Trade with the Enemy
pointed out, such pressure would take time to have an effect. In a short
war, there simply would not be any time for that to happen. By rejecting this
portion of the recommendation, the Cabinet corrected the one logical inconsis-
tency in the wartime commercial policy.
The rest of the recommendations of the Committee on Trading with the
Enemy were approved by the Cabinet. Di conseguenza, Britain’s initial wartime
commercial policy for World War I was to (1) prohibit direct trade with the en-
emy, E (2) allow indirect trade with the enemy, except in a few items that
were not permitted to leave the country because they were necessary for mili-
tary and naval means, and a few items that were not allowed to be transferred
to the neighbors of enemy countries, as these were determined to increase the
military capabilities of the enemy. No restrictions were placed on the raw ma-
terials Germany needed for its industry.
Britain’s decision to prohibit direct trade with the enemy was based on con-
cerns about public outrage at wartime trade. Using the Boer War as an analogy,
Lord Esher, a permanent member of the Committee on Imperial Defense, sug-
gested that the domestic public, and possibly allies as well, would expect max-
imum pressure to be brought to bear against Germany and would therefore be
greatly opposed to trade with the enemy: “This expectation might be formu-
lated in a demand by public opinion, voiced by the press, in such an over-
whelming force that no Government would be able to resist it.”71 To forestall
such an outcry, and the corresponding crippling pressure on the government,
the Committee proposed that, for the ªrst few weeks of war, direct trade
with the enemy be prohibited. This period of most restrictive wartime com-
mercial policy was meant to coincide with the period when the entire trans-
portation infrastructure was overloaded with transferring military personnel
and supplies to the front lines—that is, when there would have been little op-
69. “Memorandum by the Admiralty on the Economic Effect of War on German Trade,” December
12, 1908, Report, P. 388.
70. “Report,” July 30, 1912, P. 5.
71. “Note by Lord Esher,” May 2, 1912, Report, P. 427.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 39
portunity for trade to occur. So, Infatti, the prohibition was not meant to actu-
ally prohibit any trade. Later in the war, after the practical barriers to trade
were removed, the government was meant to switch to the general policy pre-
ferred by the Committee, where direct trade would be allowed. If no public
outcry materialized at the beginning of the war, the government was, likewise,
to rescind the prohibition on direct trade.72
With the exception of the prohibition on direct trade with the enemy,
Britain’s wartime commercial policy was based on weighing the economic
beneªts of trade against the military costs of potentially helping the enemy.
Trade in products that the enemy could convert into military capabilities
by the end of 1914, when the war was expected to end, was severed. Al
same time, none of the products that were of great importance to the British
economy were prohibited from trade. Infatti, despite the prohibition on direct
trade with Germany, the Board of Trade gave out numerous licenses to indi-
vidual merchants for the direct purchase of German raw materials. Such li-
censes were granted for the import of enemy goods, “which are essential for
the maintenance and development of industries in this country and cannot be
supplied to anything approaching an adequate extent from other than enemy
sources.”73 Speciªcally, a large proportion of the licenses were granted for im-
port of aniline dyes used in textile and leather industries; also prominent was
the import of various potash compounds used in the manufacture of explo-
sives, glassware, and dye-making.74
updating the wartime commercial policy
Britain’s initial wartime commercial policy was implemented at the start of the
war. But having been created for a short war, the policy quickly outlived its
usefulness. With the estimated length of the war growing, and with its inten-
sity also increasing, the commercial policy grew more stringent. The ªrst
change in policy occurred in February/March 1915, as the trench lines in
Western Europe stabilized and the expected length of the war was extended
to the end of 1915. The second change occurred at the end of 1916 and reºected
the general belief that the war could no longer be anything but exceedingly
long and bloody.
As the effective use-by date of the initial policy passed, the Committee on
72. Ibid.
73. Hubert Llewellyn Smith, “Licenses to Import Goods from Germany while the New Order in
Council Is in Force,” March 8, 1915, FO 382/186, BNA.
74. Ibid.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 40
Trading with the Enemy needed to reassess the entire strategy. In a report
printed on February 25, 1915, the Interdepartmental Committee on Trading
with the Enemy described the British policy up to that point as “concentrating
efforts mainly on preventing the enemy from receiving those articles which af-
fect his ability to prosecute the war, and to allow greater relaxation in the case
of articles of value for the civilian production alone.”75 The policy for the
short war was to prevent trade in products that were quickly converted into
military capabilities by the enemy and to allow trade in products that could
not help the enemy on the battleªeld. The report continued, “The same princi-
ple does not apply as regards the offense of trading with the enemy at the
present time.”76 Given that the war was going to last at least another full year,
the Committee decided that it now made sense to prevent certain raw materi-
als and food items from reaching the enemy. In February, numerous provisions
and victual, which may be used as food for men, were prohibited from ex-
port, as was a wide range of metals and ores.77
By the end of 1916, all the Allied strategies that the British government was
relying on to win the war had been tried and had failed. Despite Britain’s vic-
tory in the Battle of Jutland, the public perception that the war posed an exis-
tential threat to Britain did not dissipate.78 Furthermore, after two and a half
years of ªnancing the Allied war effort, Britain ªnally had to start borrowing
heavily from the United States. With the expected length and intensity of the
war growing, British decisionmakers further restricted their wartime commer-
cial policy. Import restrictions appeared and started to increase in number at a
rapid rate.79 Given the strains of war, the shipping situation became dire, forc-
ing Britain to ration tonnage on ships to ensure that products necessary for the
war economy made it to the country.80 Additionally, all licenses granted to in-
dividual merchants to trade with the enemy, regardless of the ofªcial pro-
hibitions, were suspended. After August 1916, only hosiery needles—which
were essential to the textile industry—and the occasional enemy publi-
cation were allowed to be imported from the enemy.
75. “Draft Report,” February 21, 1915.
76. Ibid.
77. More measures were added to further regulate trade with the enemy, such as certiªcates of ori-
gin and consolidated blacklists. They are not described here in the interest of conserving space.
78. Daniel Allen Butler, Distant Victory: The Battle of Jutland and the Allied Triumph in the First World
War (Portsmouth, N.H.: Greenwood, 2006), P. 94.
79. Sir Nathaniel Highmore, “Full Record of the Department," 1919, BT 11/1022, BNA, P. 34.
80. “Notes by Lord Emmott on How the Department Came into Being," 1919, BT 11/1022, BNA,
P. 4.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 41
Figura 3. Density Plot of the Distribution of Days before a Product in Each Category Was
Prohibited from Trade With the Enemy
S
T
C
tu
D
o
R
P
F
o
R
e
B
M
tu
N
substitute
good
raw
Materiale
intermediate
good
finished
good
type of product
finished good
intermediate good
raw material
substitute good
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
0
365
730
1095
1460
number of days until product is prohibited from export
The vertical line on each density curve represents the median.
trends in incremental changes in items prohibited from trade
To trace the changes discussed above in a more nuanced manner, I use my
novel dataset of all products that Britain prohibited from export during World
Guerra I. It covers 2,714 product descriptions. For each product, the initial date of
prohibition and geographic scope of prohibition is recorded, as well as any
changes that occurred during the war. Nine hundred sixty-six products were
speciªcally prohibited from trade with the enemy at some point during the
war. Additionally, I divided all products into four categories based on their rel-
ative positions in the manufacturing process—ªnished goods, intermediate
goods, raw materials, and substitute goods.
Figura 3 shows the distribution of products across the four categories. Prod-
ucts in all four categories are distributed across the full duration of the war,
which logically coheres with conversion time. Some supply chains are short
and very useful to the war, so those raw materials and ªnished goods should
be prohibited early on. Other supply chains are short but take a long time to
convert into military capabilities, so those raw materials and ªnished goods
should be prohibited further into the war.
International Security 46:1 42
A number of interesting patterns emerge from this dataset. Primo, British
decisionmakers adapted the wartime commercial policy to changes on the bat-
tleªeld. As the trenches on the Western Front solidiªed by November 1914,
and Britain saw that attrition was going to be a major feature of the current
war,81 “entrenching tools and implements,” such as pickaxes, shovels, E
spades, were prohibited from export to the enemy on November 10, 1914.82
“Machinery for trenching and digging” was not prohibited until February 3,
1915,83 when most hopes of returning to mobile warfare had faded.84 On the
same date, and following the same intuition that a war of movement was not
likely to recur, machine guns were also prohibited from trade.85 Interestingly,
machine guns were not prohibited from trade at the beginning of the war, COME
there was a wide consensus that they were useful only in wars of attrition, non
in maneuver warfare.86 Thus, carriages and mountings for machine guns were
forbidden from export at the start of the war, but not machine guns them-
selves.87 The same pattern of adapting to the changing nature of warfare can
be seen in the British understanding of a German innovation—poison gas. As
the British identiªed the compounds being used by the Germans, these chemi-
cals were prohibited from trade. Chlorine was ªrst used against the British and
French forces on April 22, 1915, and it was banned from trade on July 28,
1915.88 Phosgene was prohibited on October 19, 1915.89
Secondo, prohibitions were made based on products’ conversion times. For
esempio, harnesses and saddles used for military purposes were prohibited
from trade at the start of the war.90 The leather used to make such saddles,
81. John Turner, ed., Britain and the First World War (Crows Nest, New South Wales: Unwin
Hyman, 1988), P. 5; Ian F.W. Beckett, Ypres: The First Battle, 1914 (London: Pearson/Education,
2004); and Jeremy Black, The Great War and the Making of the Modern World (London: Continuum,
2011), P. 46.
82. London Gazette, Supplement, novembre 10, 1914, 28970, pag. 9225–9228, https://www
.thegazette.co.uk/London/issue/28970/data.pdf.
83. London Gazette, Second Supplement, Febbraio 3, 1915, 29057, pag. 1165–1169, https://www
.thegazette.co.uk/London/issue/29057/data.pdf.
84. Philip Warner, World War One: A Chronological Narrative (Leicester, UK: Brockhampton, 1999),
P. 57.
85. London Gazette, Second Supplement, Febbraio 3, 1915.
86. Frederick Victor Longstaff and Andrew Hilliard Atteridge, The Book of the Machine Gun (Lon-
don: Hugh Rees, 1917).
87. Alexander Pulling, ed., Manual of Emergency Legislation: Comprising All the Acts of Parliament,
Proclamations, Orders, Passed and Made in Consequence of the War, Supplement No. 2 to December 5th,
1914, Laws, Etc. (London: Her Majesty’s Stationery Ofªce, 1914), pag. 162–164.
88. London Gazette, Second Supplement, Luglio 28, 1915, 29244, pag. 7427–7432, https://www
.thegazette.co.uk/London/issue/29244/data.pdf.
89. Edinburgh Gazette, ottobre 22, 1915, 12864, pag. 1595–1596, https://www.thegazette.co.uk/
Edinburgh/issue/12864/data.pdf.
90. London Gazette, Second Supplement, agosto 5, 1914, 28862, pag. 6168–6169, https://www
.thegazette.co.uk/London/issue/28862/data.pdf.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 43
Tuttavia, was prohibited later on October 6, 1914.91 Metal ªttings for the har-
nesses were banned on July 28, 1915.92 Harness-making machines could be
traded with the enemy all the way up to December 18, 1917.93 Britain’s war-
time commercial policy prevented the enemy from having access to products
of immediate use on the battleªeld ªrst, then tightened progressively to pre-
vent the enemy from gaining the supplies necessary to make war-ready ªn-
ished goods. The same pattern can be traced through uniform clothing, Quale
was prohibited from trade with the enemy at the start of the war.94 “Cloth,
woolen or worsted, if suitable for uniform clothing” was not prohibited until
ottobre 19, 1914.95 “Leather, dressed or undressed, suitable for military cloth-
ing,” was prohibited on February 3, 1915.96 “Uniform clothing, second-hand
military,” which had to be taken apart and its components reused to make new
uniforms, was prohibited only on May 10, 1916.97
Figura 4 generalizes this pattern across all the products that Britain prohib-
ited from trade with the enemy. For this analysis, only those products that, at
some point during the war, were speciªcally prevented from reaching the en-
emy are considered. A product could be barred from leaving Britain because of
a domestic shortage, as well as for reasons of reducing enemy capabilities. È
possible to be conªdent about the reason for the prohibition only for those
products that, at some point, were prevented from being traded with countries
adjacent to the enemy.
As can be seen from ªgure 4, on average, it took about nine months for
ªnished goods to be prohibited from export, a year for intermediate goods, UN
year and a half for raw materials, and about two years for substitute goods.98
This pattern suggests that conversion times dictated the timing of prohibitions,
as trade in products at the beginning of the supply chain was prohibited ear-
lier than products at the end of it.
Finalmente, when the data are spatially distributed, one can see that Britain’s
wartime commercial policy varies with the decisionmakers’ expectations
91. London Gazette, Supplement, ottobre 7, 1914, 28927, P. 7999, https://www.thegazette.co.uk/
London/issue/28927/data.pdf.
92. London Gazette, Supplement, ottobre 13, 1915, 29324, pag. 10095–10096, https://www
.thegazette.co.uk/London/issue/29324/data.pdf.
93. London Gazette, Dicembre 18, 1917, 30435, pag. 13241–13242, https://www.thegazette.co.uk/
London/issue/30435/data.pdf.
94. London Gazette, Second Supplement, agosto 5, 1914.
95. London Gazette, ottobre 20, 1914, 28945, P. 8390, https://www.thegazette.co.uk/London/
issue/28945/data.pdf.
96. London Gazette, Second Supplement, Febbraio 3, 1915.
97. London Gazette, Supplement, May 10, 1916, 29574, pag. 4633–4641, https://www.thegazette
.co.uk/London/issue/29574/data.pdf.
98. All differences in means are statistically signiªcant at the .01 level.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 44
Figura 4. Average Number of Days before a Product in Each Category Was Prohibited
from Export to the Enemy
694
543
371
287
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
700
600
500
400
300
200
100
0
finished goods
(N 293)
intermediate
goods (N 323)
raw materials
(N 159)
substitute goods
(N 188)
about the length of war. Figura 5 shows the number of products that Britain
prohibited from being exported to countries adjacent to the enemy at each
month of the war through the end of 1917. While Britain was still changing its
wartime commercial policy in 1918, prohibitions in that year prevented prod-
ucts from leaving the country and were not necessarily associated with trade
with the enemy. The number of products prohibited speciªcally from
trade with the enemy remained stable after 1917.
The shaded columns in ªgure 5 represent periods when British leaders
expected a long war. The dashed line shows what a constant rate increase in
product prohibitions would have looked like. Large increases in the number of
prohibited goods tend to correlate with moments when British decisionmakers
lengthened their expectations of the duration of the war. D'altra parte,
moments of stasis, when the prohibition lists were not expanded, tend to cor-
relate with moments when the expected length of the war was fairly short.
This pattern is summarized in table 1, which shows the average number of
new prohibitions in each period of war-length expectation. Even though peri-
ods of long war expectation cover more months, the average number of prod-
ucts prohibited per month is greater than the average number prohibited
when a short war was expected. The average number of prohibitions is rather
high in the ªrst period, despite expectations of a short war, because it includes
the initial prohibitions made at the start of the conºict.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 45
Figura 5. Total Number of Products Prohibited from Export to the Enemy over Time
900
800
700
600
500
400
300
200
100
0
4
1
9
1
/
8
0
4
1
9
1
/
0
1
4
1
9
1
/
2
1
5
1
9
1
/
2
0
5
1
9
1
/
4
0
5
1
9
1
/
6
0
5
1
9
1
/
8
0
5
1
9
1
/
0
1
5
1
9
1
/
2
1
6
1
9
1
/
2
0
6
1
9
1
/
4
0
6
1
9
1
/
6
0
6
1
9
1
/
8
0
6
1
9
1
/
0
1
6
1
9
1
/
2
1
7
1
9
1
/
2
0
7
1
9
1
/
4
0
7
1
9
1
/
6
0
7
1
9
1
/
8
0
7
1
9
1
/
0
1
7
1
9
1
/
2
1
expected long war
number of products prohibited from export to the enemy
Tavolo 1. Average Number of New British Prohibitions on Trade with the Enemy, per
Period of War
8/14–10/14
11/14–2/15
3/15–5/15
6/15–10/15
11/15–1/16
2/16–12/17
short
27.0
long
20.8
short
6.3
long
30.0
short
12.0
long
16.8
By October 20, 1914, it was clear in London that neither side had managed to
outºank the other in the Race to the Sea, and that the war of movement was
over.99 The British prime minister instituted an ofªcial war council, a step that
was not deemed necessary to conduct the war earlier in the conºict. The ex-
pected end of the war was reassessed to be the end of 1915.100 As the expected
length of the war increased, so did the number of products that were prohib-
ited from trade with the enemy. The prohibited products were mostly manu-
factured metals—iron, brass, and copper.
Failing to restart a war of movement, Britain turned its attention to other
99. Turner, Britain and the First World War, P. 5; Beckett, Ypres; and Black, The Great War and the
Making of the Modern World, P. 46.
100. Warner, World War One, P. 39.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 46
theaters, searching for a way to win the war without breaking the stalemate
on the Western Front. On March 30, Prime Minister Herbert Asquith predicted
that Allied success in the Dardanelles would bring the war to an end by June
1915.101 While this expectation held, Britain’s wartime commercial policy was
not much changed. By the end of May, Tuttavia, the Dardanelles campaign
had failed; no breakthroughs appeared on the Western Front; the Germans had
signiªcant successes on the Eastern Front; and the so-called shell scandal—
the shortage of artillery shells on the front lines—was in full rage in Britain.
Asquith was forced to update his expectation of the length of the war to go
past the end of 1915.102 The Cabinet seemed a bit more pessimistic, making the
assessment in mid-June that the Allies would be unable to mount a combined
major offensive on all fronts until 1916, which meant that the war could not
end until the end of 1916, maybe even 1917.103 The pessimistic outlook was ac-
companied by a corresponding restriction in Britain’s wartime commercial
policy. Food and forage for animals as well as cotton manufactures featured
among newly prohibited items.
At the end of 1915, the Allies started planning a major offensive, which the
commander in chief of the French Armies, Joseph Joffre, had convinced them
would win the war in 1916.104 This time period is again marked largely by sta-
sis in Britain’s wartime commercial policy. The Germans seized the initiative,
Tuttavia, and attacked ªrst at Verdun in February. By April 1916, the British
government had to reevaluate their expectations of what became the Battle
of the Somme from a war-winning campaign to a diversionary battle to relieve
the pressure on the French at Verdun.105 The realization compelled the British
government to prohibit trade in tools and instruments for ªxing telegraphs,
metal-working machinery, and petroleum products not already prohibited.
In September 1916, there was some hope in Britain that the war of move-
ment would be restarted with the introduction of the tank; Tuttavia, the early
models had numerous technological problems that prevented a large portion
of them from even reaching the assault point.106 In August, Romania joined the
war effort against the Central Powers; by December 6, Tuttavia, Bucharest had
fallen, opening a new supply of grain to Germany. By the end of 1916, all strat-
egies to turn the war toward a swift conclusion had failed. Through the follow-
101. David French, British Strategy and War Aims, 1914–1916 (London: Routledge, 2014), P. 87.
102. Ibid., P. 78.
103. Ibid., P. 105.
104. Ibid., P. 112.
105. Warner, World War One, P. 95.
106. Black, The Great War and the Making of the Modern World, P. 155.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 47
ing two years of warfare, Britain’s commercial policy became increasingly
restrictive, and no moments of optimism served to decrease the expected
length of war.
Infatti, it was not until October 5, 1918, when the Allied armies started to
break through the Hindenburg Line, the German defensive position on the
Western Front, that Britain entertained any thoughts of victory.107 Of course,
this was four days after it had ªnally severed all trade with the enemy.
the role of domestic politics
A possible alternative explanation for a state’s wartime commercial policy fo-
cuses on domestic lobbying. According to this explanation, the decisions about
which products were prohibited from trade and which were allowed would
stem from the strength of the lobbying efforts of individual industries relying
on trade in those products. When making decisions about which items were to
be prohibited from indirect trade with Germany throughout the war, Tuttavia,
the British government was largely unaffected by domestic lobbying. Al-
though such lobbying was strong, Britain’s wartime commercial policy contin-
ued to be made for security and economic reasons, not because of pressure
from domestic commercial interests. This is not to say that British business in-
terests had no effect at all on policy. On occasion, when industry representa-
tives could prove that a speciªc product description could be used to separate
products that had military applications from those that did not, the prohibition
lists were amended with the more targeted product descriptions.108
An illustrative example is the effort made by Britain’s domestic ªrms to re-
move “lead, in all forms” from the list of items prohibited from export from
the country. In a letter dated August 14, 1914, the director of Locke Lancaster, UN
prominent lead merchant, argued that the product description was too inclu-
sive.109 It prohibited, for instance, the export of tea lead linings for tea chests
and white lead, neither of which has a military application, but both of which
are necessary to sustain exports of certain industries. It was also advocated
that pig lead, which was acknowledged as having military applications,
107. Spencer C. Tucker, The Great War, 1914–18 (Bloomington: Indiana University Press, 1998),
P. 171.
108. John McDermott, “‘A Needless Sacriªce’: British Businessmen and Business As Usual in the
First World War,” Albion: A Quarterly Journal Concerned with British Studies, Vol. 21, No. 2 (1989),
P. 270, doi.org/10.2307/4049929.
109. Letter from Director of Locke Lancaster to Assistant Secretary, Commercial Department,
Board of Trade, agosto 14, 1914, HO 45/10728/254895, NO. 49, BNA. See also follow-up conversa-
zione, “Re. Prohibition of the Export of Pig Lead, Sheet and Pipe,” HO 45/10728/254895, NO. 279,
BNA.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 48
should be prohibited from export to the enemy, not restricted from leaving the
country altogether. This request was justiªed by the claim that both France and
Russia had previously obtained their pig lead from Germany, and if Britain
was not available to supply it, the allies would be forced to turn toward pur-
chasing this necessary article from the enemy. In response, on August 20, 1914,
the prohibition on “lead, in all forms” was replaced with “lead, pig, sheet, O
pipe”; despite pressure from lead exporters, Tuttavia, this item was still pro-
hibited from leaving the country.110
The only clear case of success for domestic lobbying was a delegation repre-
senting 70 percent of London’s confectionary trade, which was able to con-
vince the Board of Trade to remove “confections, all kinds” from the list
of items prohibited from export. Trade in confections was prohibited from
agosto 5 to August 20, 1914.111 This product did not reappear on the list of
prohibited items until March 12, 1917; it was removed again on May 11,
1917.112 Such leniency was given to the group of conferences because its mem-
bers were able to show convincingly that there was zero home demand for
their products and that their exclusive clientele were British colonies.113
D'altra parte, rather unsurprisingly, merchant requests to al-
low the export of ammunition, even to African countries, were denied by
the government.114
Conclusione
A state’s decision to prohibit or allow trade with the enemy in wartime is effec-
tively a balancing act between military and economic considerations. On the
one hand, the state seeks to avoid helping opponents increase their military ca-
pabilities. D'altra parte, severing trade reduces that state’s revenue,
which inºuences its long-term security. The goal of wartime commercial policy
is to maximize a state’s revenue from trade while minimizing the negative mil-
itary consequences of that trade. States navigate that trade-off with the help of
two calculations. Primo, to determine whether continued trade will help the en-
110. Pulling, Manual of Emergency Legislation, pag. 168–170.
111. London Gazette, Second Supplement, agosto 5; and London Gazette, agosto 21, 1914, 28876,
pag. 6583–6584, https://www.thegazette.co.uk/London/issue/28876/data.pdf.
112. London Gazette, Marzo 13, 1917, 29982, pag. 2510–2511, https://www.thegazette.co.uk/
London/issue/29982/data.pdf.
113. “Papers Relating to the Proclamation of August 20th Concerning the Prohibition of Exporta-
tion of Certain Articles,” August 20, 1914, BT 11/7, BNA, P. 7397.
114. Letter from African Lakes Corporation Limited to Privy Council Ofªce, novembre 18, 1914,
PC 8/789, NO. 114676, BNA.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 49
emy in the current war, the state compares the amount of time it takes the
enemy to convert each traded product into some military capability with
the expected length of the war. Secondo, to establish if trade can be safely sev-
ered in wartime, the state compares the revenue generated by each product to
what it needs to invest to guarantee its long-term security. Trade that can be
cut off based on revenue considerations and that also can help improve the ad-
versary’s military capabilities is likely to be prohibited in wartime.
This theory has important implications for contemporary international
politics—speciªcally, the question of whether China can rise peacefully. Given
the return of great power politics with the recent transition from unipolarity to
multipolarity, the central question concerning policymakers and scholars alike
is whether China’s impressive economic growth will lead to military conºict
with its neighbors, O, more importantly, war with the United States.
One of the most prominent arguments for optimism regarding China’s rise
is based on economic interdependence theory.115 Given the extensive trade
links between China and its neighbors as well as with the United States, Tutto
those states have a powerful incentive not to ªght with each other, because
war would put an end to trade between the combatants, thus undermining
their prosperity at home. All else being equal, trade acts as a deterrent to war.
Ovviamente, this deterrent exists only if trade relations between wartime ene-
mies are severed. Wartime trade between adversaries is not only possible,
Tuttavia; it is quite prevalent in the historical record, as states have good rea-
sons to keep trading with their enemies even during war. But if speciªc cir-
cumstances cause states to trade during wartime, trade cannot deter such a
military conºict.
Turning to the case of a rising China, there are three types of wars that it
could ªght: war with another nuclear-armed power; war with a close ally of
the United States; or an asymmetric war, where it engages against a much
weaker country without a strong patron. Only with regard to an asymmetric
war is trade likely to act as a deterrent, although, even then, its impact on war
initiation is questionable.
There are two main scenarios where China might become involved in a war
with another nuclear-armed power: a conºict between China and the United
115. Michael Mousseau, “The End of War: How a Robust Marketplace and Liberal Hegemony Are
Leading to Perpetual World Peace,” International Security, Vol. 44, No. 1 (Estate 2019), pag. 160–
196, doi.org/10.1162/isec_a_00352; and Aaron L. Friedberg, “The Future of U.S.-China Relations:
Is Conºict Inevitable?” International Security, Vol. 30, No. 2 (Autunno 2005), pag. 7–45, doi.org/10.1162/
016228805775124589.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 50
States caused mainly by geopolitical competition;116 and a war between China
and India that grows out of their long-standing border dispute.117 A war be-
tween two nuclear-armed states is likely to be short and limited. They cannot
ªght a conventional, total war such as World War I or World War II given the
presence of nuclear weapons. After all, an attack that threatens the survival of
a nuclear state is likely to trigger nuclear retaliation.118 Thus, wars of conquest
or unconditional surrender are not viable options. Even a limited war between
two nuclear-armed powers is unlikely to last long, because of the risk of nu-
clear escalation. Both sides would face powerful incentives to negotiate a set-
tlement to end the conºict. In limited wars of short duration, Tuttavia, Tutto
belligerents are likely to permit considerable amounts of trade, mainly because
the enemy would not have sufªcient time to convert the gains from trade into
military capabilities. Così, there would be no beneªt from severing trade.
Given the expectation of continued trade in a war between two nuclear-armed
powers, trade is unlikely to deter such a conºict in the ªrst place. These dy-
namics were at play, albeit on a smaller scale, in the recent border skirmishes
between China and India. Despite their extensive economic relations, China
did not hesitate to engage in aggressive behavior toward India. Inoltre,
neither side seriously restricted trade during their conºict.119
The second type of war that a rising China might ªght is with a close ally
of the United States, such as Japan, Taiwan, or South Korea. If a war broke out
between China and a close U.S. alleato, it is highly likely that the United States
would become directly involved in that conºict. The United States has a large
116. Graham Allison, Destined for War: Can America and China Escape Thucydides’s Trap? (New York:
Houghton Mifºin Harcourt, 2017); Adam P. Liff and G. John Ikenberry, “Racing toward Tragedy?
China’s Rise, Military Competition in the Asia Paciªc, and the Security Dilemma,” Interna-
tional Security, Vol. 39, No. 2 (Autunno 2014), pag. 52–91, doi.org/10.1162/ISEC_a_00176; Tommaso J.
Christensen, “Fostering Stability or Creating a Monster? The Rise of China and U.S. Policy toward
East Asia,” International Security, Vol. 31, No. 1 (2006), pag. 81–126, doi.org/10.1162/isec.2006
.31.1.81; and Mearsheimer, The Tragedy of Great Power Politics.
117. Sumit Ganguly and Manjeet S. Pardesi, “Why We Should Worry about China and India’s Bor-
der Skirmishes,” Foreign Policy, May 23, 2020, https://foreignpolicy.com/2020/05/23/india-china-
border-skirmishes/; Ameya Pratap Singh, “Why Another Sino-Indian War Is Unlikely,” Diplomat,
Giugno 1, 2020, https://thediplomat.com/2020/06/why-another-sino-indian-war-is-unlikely/; E
Kevin Brown, “Who Would Be Favored in a China-India Conºict in the Himalayas?” National In-
terest, ottobre 23, 2018, https://nationalinterest.org/feature/who-would-be-favored-china-india-
conºict-himalayas-34167.
118. Robert Jervis, The Meaning of the Nuclear Revolution: Statecraft and the Prospect of Armageddon
(Ithaca, N.Y.: Cornell University Press, 1989); and Caitlin Talmadge, “Would China Go Nuclear?
Assessing the Risk of Chinese Nuclear Escalation in a Conventional War with the United States,"
International Security, Vol. 41, No. 4 (Primavera 2017), pag. 50–92, doi.org/10.1162/ISEC_a_00274.
119. Banikinkar Pattanayak, “Border Clash Fails to Dampen India-China Trade,” Financial Express,
settembre 9, 2020, https://www.ªnancialexpress.com/economy/border-clash-fails-to-dampen-
india-china-trade/2078240/.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Wartime Commercial Policy 51
military presence in both Japan and South Korea and is committed to defend-
ing them if they are attacked by China.120 U.S. policymakers also have made it
clear that they would defend Taiwan if China attacked. In essence, the pros-
pect of direct and immediate U.S. involvement in this second type of conºict
makes its dynamics similar to a war between two nuclear-armed great powers.
Given the dangers of nuclear escalation in such a war, it would likely be short
and limited, which means that wartime trade would be allowed. Così, trade is
not likely to deter a war between China and a close U.S. alleato.
China could also get involved in an asymmetric war where it ªghts a mili-
tarily inferior country that is not closely allied with the United States. Vietnam
and Indonesia are possible examples. It is only in this type of war that trade
would likely act as a deterrent to the start of a war; although its ultimate im-
pact would be questionable. Speciªcally, from the weaker state’s perspective,
an asymmetric war against a much stronger opponent would likely be a pro-
tracted guerrilla war.121 The weaker state would have strong incentives to
sever trade to increase the costs of war imposed on its opponent. With no war-
time trade between the belligerents, the conditions for economic interdepen-
dence theory hold. The prospect of losing trade with China could dissuade a
weaker opponent from starting such a war. Allo stesso tempo, Tuttavia, there
are many other reasons why the weaker state might be deterred from initiating
such a conºict—for example, the superior capabilities of the opponent.
From China’s perspective, an asymmetric war against a weaker opponent
would likely be quick and easy. China would have little reason to sever trade
during the war, but it would be aware that its opponent would have strong in-
centives to cut off trade with China. The prospect of losing trade with the
weaker state could theoretically dissuade China from starting the war. Ma,
given the trade asymmetries between China and its much weaker neighbors, Esso
is unlikely that the costs of losing such trade would be particularly high. In
short, it is only with the prospect of an asymmetric war that trade might serve
as a deterrent to conºict, although its ability to actually accomplish this goal
is suspect.
Overall, the expectation that signiªcant economic ties will help prevent
conºict between a rising China and its neighbors or the United States is mis-
guided. Inoltre, the greater the economic interdependence between China
and any of its potential foes, the more likely it is that they will trade during a
120. Victor D. Cha, “Powerplay: Origins of the U.S. Alliance System in Asia,” International Secu-
rity, Vol. 34, No. 3 (Inverno 2009/10), pag. 158–196, doi.org/10.1162/isec.2010.34.3.158.
121. Arreguín-Toft, “How the Weak Win Wars.”
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
International Security 46:1 52
war, thus eliminating trade considerations as a deterrent to military conºict.
When two economies are tightly integrated, certain inputs from each state are
necessary to the value-added manufacturing supply chains of the other state.
If a product is essential to a state’s economy and only the enemy belligerent
can provide it, the state is more likely to continue trade in that product even if
this trade helps the enemy ªght the war. Così, closer integration of two econo-
mies does not deter conºict; Infatti, that interdependence makes continued
trade in the event of war more likely.
Finalmente, my theory has implications for the policy of decoupling, which has
attracted much attention in part because of the fear that the United States is too
dependent on raw materials from China. Some U.S. policymakers have re-
cently advocated severing strategic trade with China to reduce the potential
risks to the United States. They argue that relying on China for raw materials
gives Beijing signiªcant leverage over Washington, especially because China
could sever this trade at a crucial moment in a war and put the United States at
a signiªcant disadvantage.122 Such fears, Tuttavia, are unfounded. Raw mate-
rials tend to have long conversion times, as they must be shipped from China
to the United States, go through an entire production chain to be converted
into something militarily useful, and shipped to the battleªeld. Because raw
materials tend to have lengthy conversion times and a potential U.S.-China
war would likely be short, trade in raw materials is likely to continue during
conºict. Neither side would want to bear the costs of severing trade when such
a move would have hardly any military beneªts.
122. Jonathan D. Pollack and Jeffrey A. Bader, Looking before We Leap: Weighing the Risks of US-
China Disengagement (Washington, D.C.: Brookings Institution Press, Luglio 2019), https://www
.brookings.edu/wp-content/uploads/2019/07/FP_20190716_us_china_pollack_bader.pdf.
l
D
o
w
N
o
UN
D
e
D
F
R
o
M
H
T
T
P
:
/
/
D
io
R
e
C
T
.
M
io
T
.
e
D
tu
/
io
S
e
C
/
UN
R
T
io
C
e
–
P
D
l
F
/
/
/
/
4
6
1
9
2
0
7
9
8
5
6
/
io
S
e
C
_
UN
_
0
0
4
1
2
P
D
.
F
B
sì
G
tu
e
S
T
T
o
N
0
8
S
e
P
e
M
B
e
R
2
0
2
3
Scarica il pdf