I’ve recently read a widely-noted new book by Graham Allison, Destined for war: Can America and China escape Thucydides’s trap? (Boston: Houghton Mifflin Harcourt). The question that Allison addresses, whether the United States and People’ s Republic of China (PRC) might somehow go to war, is of supreme importance and I want to set down my initial thoughts in response to his treatment.
Graham Allison will long be remembered as a scholar. His 1971 book, Essence of Decision: Explaining the Cuban Missile Crisis, is of enduring interest, not so much for its account of the history of the crisis as for its analysis of alternative models of decision making.
In light of this I approached this book with positive expectations. But sadly, Destined for war, I believe, is not of that caliber. In fact, in many ways it seems quite peculiar. Fortunately, the best parts are the very final conclusions and recommendations. Less fortunately, the path to them is very frequently a matter of the proof of the obvious by means of the dubious.
Unless you seriously doubt that China is economically strong and growing stronger, or that the possibility over the next few decades of conflict between China and the United States can be disregarded altogether because it is less, say, than one in a million, then you will find nothing in the first three-quarters of this book that is worthy of your time and attention. If you read it and find that it contains novel and important information this will largely be because you have been misled.
I will explain.
Essentially, Allison’s arguments are
- China is surging ahead of the United States both economically and militarily, so much so that nothing can stop it from achieving global dominance.
- This situation of a rising power (China) overtaking a formerly dominant one (us) is very dangerous. It has frequently led to war in the past and could very well do so now.
He develops these points at some length, but in an easy, non-academic style. So non-academic, indeed, that in a number of areas he plays pretty fast and loose with the facts. Although Allison wants us to believe that he is dispassionately examining the issues the reality is that there is much more exhortation than analysis here.
One aspect of the problem is that, learned as he is in some areas, Allison is a political scientist specializing in IR (international relations) theory and expert neither in economics nor Asian history, nor has he more than very distinctly limited understanding of technology and military affairs. Sometimes he shows astute judgement in which sources to consult in these matters, but too often he doesn’t.
His weaknesses in economics are on full display in his projections about the future relative economic strengths of the U.S. and China. The one thing we can be quite certain of about the economic future is that it’s highly uncertain. To be sure, there are things in which momentum or accumulated stocks play a very big role, permitting a degree of fairly confident prediction; population, for instance, or the size of naval fleets. But in things like GDP growth, technological innovation, or social trends, for instance, we are flying blind and a look back at the confident predictions of experts in the past is (or certainly should be) quite sobering.
Instead Allison takes at face value PRC pledges of 6.5%/year growth for a number of years to come, contrasting this with the U.S. trend of little more than 2%/year, leading him to assert confidently that China will soon vastly outstrip the United States.
For the United States the risks are largely in policy. Immigration is a significant plus factor and if Trump Administration plans to cut it back severely go through there will almost surely be reductions to growth in both GDP and GDP per capita. In addition, some Republican officials have repeatedly indicated a real willingness to force a default on U.S. debt as a means to force through policy changes. Most economists believe that any real default (lasting more than a very brief period) could cause a worldwide financial collapse comparable to that of 2008, and leave a lasting residue of poor U.S. credit and high risk premia on U.S. debt, all of which could be expected to have severe impacts on economic growth.
The PRC leadership seems unlikely to stumble into such ruinous policy choices, but China’s economic risks are nevertheless serious. Sustained rapid growth has led to accumulation of imbalances and distortions in the economy. The government has proven adept and firm about managing the economy, but some problems are growing to levels that threaten to defy even the best management. (It’s interesting to note that many people who insist that government destroys everything it touches in the U.S. display a great deal of faith in the PRC government’s ability to manage the country’s economy.)
But China is well into the most dramatic example of the demographic transition ever seen, reflecting a confluence of natural forces associated with economic modernization exacerbated by the “one-child policy” the government insisted upon, with some exceptions, from 1979 to 2015. As a result the country’s working-age (and military-age) population will decline steadily and its old-age population will grow for decades to come, and its total population will start to shrink. Unless there is compensating strong growth in labor productivity this will result not only in rising social expenses but slowly rising GDP.
The United States by contrast enjoys higher effective reproduction rates (due largely to immigrant groups) as well as population growth due to immigration, which combine to provide a powerful support for GDP. While the U.S. too has an aging population problem is it not nearly as severe or progressive as China’s, and the influx of young immigrants (for as long as it’s allowed to last) provides a pool of people to work in caring for the old.
Allison throws out numbers in a fairly unsystematic way. He expresses projections in constant U.S. dollars (baseline usually not specified) at PPP, asserting confidently that PPP is the best way to compare national incomes for defense purposes. It’s widely agreed that PPP is best for comparing overall welfare, but that’s because it’s constructed for that purpose. It should be self-evident to anyone who thinks about it that this in general is not the best way to make military-potential comparisons. Just because a dollar spent in China will buy twice as many (say) haircuts or restaurant meals as it will in the United States doesn’t mean that it will buy twice as much military capability. Allison truncates a CIA statement in a way that significantly distorts its meaning, changing
The data derived from the PPP method probably provide the best available starting point for comparisons of economic strength and well-being between countries. … GDP derived using the OER [official exchange rate] method should be used for the purpose of calculating the share of items such as exports, imports, military expenditures, external debt, or the current account balance, because the dollar values presented in the [CIA] Factbook for these items have been converted at official exchange rates, not at PPP. One should use the OER GDP figure to calculate the proportion of, say, Chinese defense expenditures in GDP, because that share will be the same as one calculated in local currency units.
PPP “provides [sic] the best available starting point for comparisons of economic strength and well-being between countries.”
On his page 7 Allison casually suggests that China might double or even quadruple its levels of labor productivity over the next ten to twenty years. When I’ve put this to economists familiar with such matters I’ve always gotten the same reaction: a snort of laughter. As one says, if Allison knows the secret of productivity growth in a maturing economy (where services are rapidly displacing manufacturing as the principal employer) it would be nice to share. It’s reasonable to expect more rapid productivity growth in China than in the U.S. (since China can copy us in an effort to catch up). But the major obstacles to productivity in China are structural and at best will yield slowly and painfully to reform efforts. Corruption is clearly a very major problem and Allison speaks in upbeat tones of Xi Jinping’s drive to extirpate it. Yet Chinese regimes have been inveighing against corruption time out of mind and have periodically mounted draconian campaigns to sanction the malefactors with little fundamental effect. The social roots underlying corruption, known as guanxi, reach to the very heart of Chinese society — and the ruling Chinese Communist Party (CCP) that Xi heads. The CCP has waged war against guanxi since long before it gained power, but with quite limited success. Guanxi is changing but we will not soon see a China with even U.S.-levels of corruption, let alone those of Germany, the Netherlands, or the U.K. There can be little doubt that corruption will remain a major drag on Chinese economic efficiency for a long time to come.
For what it’s worth here are the OECD’s projections of GDP not just for China and the U.S. but for all of the large and middling powers active in the region, in 2005 U.S. dollars at PPP, out three decades in the future:
|Projections of GDP at PPP in 2005 U.S. dollars, trillions.|
Note that China has no allies other than North Korea (whose economy is too small to show up in this table). The policy of the Trump Administration seems to be to weaken U.S. alliances but if we assume that this does not become permanent the United States could reasonably be expected to be joined by others in any conflict. In the table it’s easy to see how the United States together with plausible allies might outweigh China in GDP measured at PPP.
To further buttress his case Allison summons a number of more or less striking anecdotes about Chinese economic expansion and some deep-sounding generalizations regarding Chinese historical trends. The anecdotes are all very positive, but anyone who has followed China’s development at all closely will know that it would be equally easy to assemble a parade of disasters and horror stories. It’s not impossible that China has at last found the clear path to unending successes without setbacks, but it’s not something to bet a great deal on, I should think.
Periodically through the book Allison tells the reader tales of the magnificence and uniqueness of classical Chinese civilization, including that it is 5,000 years old and continuous since then, and that until the rude intrusion of Britain followed by others in the early- to mid-1800s, China had only dealt with other nations on a lofty patron-client basis. While this is all a standard part of the PRC’s patriotic myth, taught to every Chinese schoolchild, it takes very little scholarly reading to discover the limitations and distortions in it. For the regime, of course, these myths serve an understandable purpose, but their value for Allison is less clear. Naturally their falsity tends to diminish the impact of the “lessons” he derives regarding Chinese history.
The distortions of Allison’s view of China may derive, I suspect, from his friendship and admiration for the late Lee Kuan Yew (1923-2015), long the strongman autocrat of Singapore. Lee was part of the Chinese diaspora, his great-grandfather having emigrated in mid-1800s. But he was educated at Cambridge and practiced as a barrister before leading Singapore’s successful movement to persuade the British to bestow independence on a city that had actually been created by Britain in the first place. Over the next few decades Lee oversaw the tremendous increase in wealth in Singapore.
Allison, nearly a quarter-century Lee’s junior, wrote a very admiring book entitled Lee Kuan Yew: The Grand Master’s Insights on China, the United States, and the World. There’s no question that Lee was an exceptional man who had a good deal of reason and opportunity to study the China of his time, but of course there were limitations to his knowledge, and he had a particular perspective that may not serve all purposes. Nevertheless, Allison seems to have adopted Lee as his one authority on China except for occasional references to Henry Kissinger (who also has demonstrated limits to his understanding of Chinese history).
Allison is surely right that China has been growing at unprecedented rates, economically, for the past three decades, much as Japan did in the 1960s through the 1980s. When you start from a base as low as China’s it is possible to catch up at astonishing rates — for a while. But just as Japan did China faces some factors which can limit its rate of growth. Which ones it will encounter first and how severe they will be remains to be seen, but one thing we know without question is that high rates of exponential growth cannot continue forever.
Allison also warns that China will overtake the United States in science and technology (S&T). The issues are highly complex and Allison is scarcely the first political scientist to issue naïve summary judgements. There is a great deal that is not known about S&T, including how best to stimulate its growth. The Soviets devoted great attention and massive resources to it for decades, with spotty success. Scotland, a small country with relatively modest wealth, has been a major source for 2½ centuries. There are many hypotheses and not enough evidence to pass reasoned judgement on most of them.
One thing that is certainly clear is that over the longer run technology depends on a supply of scientific knowledge to permit significant advances. The United States throughout its history has never lacked for those ready and able to use the scientific knowledge of the day to create technological knowledge. But the U.S. record in consistently producing scientific knowledge has been much more uneven. With many of our brightest finding Wall Street more alluring than the laboratory and an increasingly anti-intellectual political establishment imposing arbitrary and ignorant restrictions on research while cutting funding there certainly are real clouds on our scientific horizon. It seems all too possible that we are eating our seed corn.
In addressing the bottom line in the balance of power Allison again quotes Lee Kuan Yew, who never fought a war or studied war, to the effect that economic power outweighs military power. He seems unaware of the extent to which the Allies used their maritime power advantage to destroy German economic power in World War I, and Japanese as well as German power in World War II. China’s economic power is vulnerable in very much the same way as Germany’s was. In any serious conflict with the maritime West her overseas trade would stop almost instantly as warships seized or if necessary sank ships bound to or from Chinese ports; what then of her economic power?
Finally, Allison speaks with rather cynical admiration of China’s “unique mastery in using hard instruments of ‘soft power’ ” in employing market power to damage trading partners in retaliation for political affronts. He gives as a major example China’s actions to withhold rare-earth metals supplies from Japan in 2010 in order to deliver a rebuke for the seizure of a Chinese fishing vessel. He fails to point out the costly longer-term damage this has done to China’s markets for its rare-earth metals as well as the caution it has engendered among those considering importing any critical goods from Chinese suppliers. Western firms regularly assess supply-chain vulnerabilities and are particularly wary about depending on Chinese sources. Such are the rewards of heavy-handed trade policies.
Allison also weighs in on the military balance, seeing China pulling ahead as the U.S. stagnates or declines. He shows here his lack of knowledge and understanding of military technology and capabilities and cites weak sources (virtually the only kind publically available, of course). The reality is a great deal more nuanced and complex. The PRC is moving ahead militarily but there seems no visible prospect that it will be able to seriously erode U.S. capabilities to do it very great damage in a conflict. While Chinese forces may increasingly be able to put U.S. forces at risk when operating in certain areas there is no way they can prevent them from continuing to exercise overall dominance over the ocean commerce that is essential for Chinese prosperity nor from maintaining robust support for those states in the region that do not wish to submit to Chinese control. China is inherently a land power and probably could, if it chose, come to dominate quite a large swath of the Eurasian continent, but it faces very different and more formidable problems in challenging so great a maritime power as the United States is.
In the second section of the book Allison starts with his analysis of the lessons to be gained from Thucydides’ (c.460BCE-c.395BCE) magnificent history of the Peloponnesian War (431BCE-404BCE), in which Athens and Sparta alike were ruined. He keynotes this with two quotations from Thucydides
The final point was reached when Athenian strength attained a peak plain for all to see and the Athenians began to encroach upon Sparta’s allies. It was at this point that Sparta felt its position was no longer tolerable and decided by starting this present war to employ all energies in attacking and, if possible, destroying the power of Athens. —Thucydides, History of the Peloponnesian War
It was the rise of Athens and the fear that this instilled in Sparta that made war inevitable. —Thucydides, History of the Peloponnesian War
Allison sees this as an instance of a general tendency for tensions and potentially conflict to arise out of a dynamic power (such as Athens was) to challenge the position of an established dominant power (e.g., Sparta), terming it the Thucydides trap.
IR specialist that he is, Allison has conducted a statistical examination, compiling a file of 16 cases which he sees as fulfilling the requirements of his Thucydides trap concept and showing that in 12 of them a war resulted. Here he explores the original Greek case in modest depth, based on Thucydides’ book (which is essentially the only source we have). Then he very briefly sketches five selected cases from the 17th to the 20th centuries, drawing sweeping conclusions about why each resulted in war. Finally he presents the case of the First World War, concluding that it too is a clear-cut example of the action of the Thucydides trap.
I’m going to focus on two cases which I happen to know something about — a good deal more about them, as it turns out, than Allison seems to.
In the case of the war between Japan and the United States, the Pacific War of 1941-1945, Allison treats it as a separate conflict in isolation from the other currents of World War II. But from such a false starting assumption, as Gerhard Weinberg has shown, it is possible only to reach erroneous conclusions, which Allison proceeds to do. In fact, at its roots, the Pacific War resulted not from Japan’s rise (which amounted to little at that time) but from Hitler’s. It was by allying itself with Hitler and tying its fate to his that it brought war on its head. The Pacific War was no instance of the Thucydides trap at all but rather one element of a much larger and more complex conflict.
Allison is far from alone in muddled thinking about the origins of the Pacific War, and there are many authors whose treatments are little better, so perhaps we can pass this over as an aberration — or at least we might were it not paired with a much less forgivably distorted account of the origins of World War I.
In Allison’s telling, the Anglo-German battleship-building race of 1898-1912 and the fears it evoked in Britain’s rulers and public was at the root of the conflict; everything else is secondary. He’s not the first person to advance such a view but it is not at all widely accepted. For his purposes, however, this serves to convert what most present as a multi-party, multi-origin conflict into a straightforward case that can be fitted into his Thucydides trap framework; but it certainly fails to impress me.
Two years before this book was published Allison contributed a chapter stating some of its main arguments to a volume edited by two other noted scholars and including contributions by a number of others, whose theme is summarized in its title: The next great war? The roots of World War I and the risk of U.S.-China conflict. Several of the other contributors to The next great war also are big names in IR or Harvard colleagues who are eminent in other fields. And yet The next great war receives no citations in Allison’s new work, which ignores many of its themes. It certainly seems odd at best.
“Beware, be very aware!” is the theme of Allison’s third section. He warns that China is ten feet tall and growing, and that her leaders are supremely wise, patient, and calculating, while America’s are naïve louts who are putty in their masterful hands. Then he contrives several scenarios which he deploys to argue that they might easily be drawn into hasty and fantastically ill-considered decisions for war over trivial causes, with no gain whatever in prospect. All in all it’s a remarkable performance, all the more so from a man justly famous for his subtle and realistic analysis of the foibles and strengths of high-level decision-making.
Throughout he displays profound ignorance of U.S. military and defense technological capabilities and potentials, as well as a consistent tendency to take Chinese propaganda about Chinese capabilities at full face value. This sort of tendency to make sweeping net assessments founded in deep ignorance of the realities has historically been quite dangerous. It unquestionably contributed to the origins and both world wars, and to needless defeats and suffering.
The final, relatively brief section focuses on policies and practices to ameliorate the risks of conflict. Following as it does after 184 pages of frequently meretricious arguments based in often ill-founded factual claims, it comes as considerable relief to find that for the most part its recommendations are sensible and the claims made for them are modest.
It’s all quite odd, extremely so. Most people, I find, are entirely prepared to take it for granted that a great power with a very prominent chip on its shoulder, as China has, presents some risk of war that we must be wary of. Surely it doesn’t require 184 pages — more than three-quarters of this book — to establish that China is indeed a great power with a chip on its shoulder, and to point out that its growth presents challenges.
In the process it might be well to dispose of the two principal “impossibility arguments” that people raise in an effort to deny the threat:
- War is “impossible” between two nuclear powers, except in the form of a catastrophic nuclear exchange.
- War is “impossible” between two states that depend on trade between them for much of their prosperity.
Both are addressed very briefly in the final section, but not in the depth needed and never before the end.
It is well established (as has long been apparent to sellers of used books) that of buyers of blockbuster bestseller books, only a minority ever read as far at the halfway point. It’s easy to believe in this case, where the first 75% of the book is largely dross. Readers would be well served in this book by starting it on page 187. Ignoring the first parts would give them the added benefit of avoiding contact with much toxic intellectual waste to be found in its earlier sections.
To give so bad a report of a book by an author whom I respect and admire for his previous work gives me no pleasure, but I don’t feel I have much choice.
 A 1999 revision coauthored with Philip Zelikow updates the history in light of much added evidence; its title is unchanged.
 Regarding the unreliability of economic forecasts, even in the relatively short term, see Hites Ahir and Prakash Loungani, “There will be growth in the spring: How well do economists predict turning points,” CEPR, http://voxeu.org/article/predicting-economic-turning-points; Prakash Loungani, “How accurate are private sector forecasts? Cross-country evidence from consensus forecasts of output growth,” International Journal of Forecasting 17, no. 3 (2001); Kevin J. Lansing and Benjamin Pyle, “Persistent overoptimism about economic growth,” FRBSF Economic Letter 3 (2015).
 For a different view as well as independent estimates see Peter E. Robertson and Adrian Sin, “Measuring hard power: China’s economic growth and military capacity,” Defence and Peace Economics 28, no. 1 (2016).
 Allison pp. 10-11.
 For a survey of guanxi and related issues see Thomas B. Gold, Doug Guthrie and David L. Wank, eds., Social connections in China: Institutions, Culture, and the Changing Nature of Guanxi (Cambridge: Cambridge University Press, 2002).
 Regarding the discontinuities in the 4,000 year history of Chinese civilization (not 5,000 years as Allison repeatedly insists) see John Keay, China: A History (London: Harper Press, 2009), or any other history informed by the discoveries of the past three decades. For the realities of shifts in China’s relations with the world as circumstances changed see Morris Rossabi, ed., China among equals: The middle kingdom and its neighbors, 10th-14th centuries (Berkeley: University of California Press, 1983).
 For a somewhat better informed debate about the evolution of critical defense capabilities see Stephen Biddle and Ivan Oelrich, “Future Warfare in the Western Pacific: Chinese Antiaccess/Area Denial, U.S. AirSea Battle, and Command of the Commons in East Asia,” International Security 41, no. 1 (2016) together with Andrew S. Erickson et al., “Correspondence: How Good Are China’s Antiaccess/Area-Denial Capabilities?,” International Security 41, no. 4 (2017).
 See Gerhard L. Weinberg, Germany, Hitler, and World War II (Cambridge: Cambridge University Press, 1995), pp. 205-06 as well as Gerhard L. Weinberg, “Grand Strategy in the Pacific War,” in Pearl to V-J Day: World War II in the Pacific, ed. Jacob Neufeld, William T. Y’Blood and Mary L. Jefferson (Washington: Air Force History and Museums Program, 2000), pp. 1-3.
 For well-regarded recent treatments of the war’s origins see Christopher Clark, The Sleepwalkers: How Europe Went to War in 1914 (New York: Harper, 2013) as well as Margaret MacMillan, The War That Ended Peace: The Road to 1914 (New York: Random House, 2013).