Ships over Troubled Waters: Examining Naval Development in Asia

Abstract

What accounts for the growth of naval power in Asia, and are we witnessing a new naval arms race? Through a unique quantitative assessment of naval development in Asia, I show that economic and geostrategic factors are driving naval development in the region. Additionally, I examine the growth of naval might in China, Japan, India, and South Korea, and find evidence of a potential naval arms race in Asia. The article’s findings point to the importance of testing for regional variation in naval development and using specific weapons systems to test for evidence of an arms race.

Keywords

Naval power arms race naval development Asian region Asian rivals

A new age of naval growth is dawning in Asia. China is reclaiming its maritime heritage lost in the aftermath of the First Sino-Japanese War. China’s China’s great rival, Japan, is also rebuilding the naval might they it lost in the aftermath of World War II as they it faces shifting regional security challenges. Furthermore, we are witnessing the emergence of India and South Korea emergence as major naval powers. If the twenty-first21st century is shaping up to be a century of growing naval power, scholars will most certainly see it as an Asian century.

What accounts for the growth of naval power in Asia, and are we witnessing a new naval arms race? Although there is a growing amount of quantitative work focusing on naval development, it often avoids focusing on specific regions. Instead, there is an assumption that factors that have an impact on naval development operate similarly in all regions. Here, I expand on the Crisher and Souva (2014) naval data set (by including amphibious warships) to present a unique quantitative assessment of naval development in Asia. At the regional level, factors such as energy consumption and GDP are playing a crucial role in naval development. Geostrategic factors such as being situated by a naval chokepoint are also playing a role, although there is an inconsistent relationship between economic competition and naval development.

An additional contribution made by this article is its examination of whether there is an ongoing naval arms race in Asia. Although there is a robust literature on arms races, it tends to emphasize military expenditure rather than specific military platforms (Stoll, 2017). As such, there has been little quantitative work that focuses on naval arms races. Focusing on naval tonnage, I find evidence of a potential naval arms race in Asia, especially in the post-Cold War era. There is particularly strong evidence that the Chinese and Japanese navies are growing in response to the growth of rival navies.¹ India, however, seems to be immune from this development in the post-Cold War era.

The article proceeds as follows. First, I discuss the trends of naval growth in Asia versus the rest of the world. Second, I discuss various factors that could have an impact on naval growth and conduct an empirical analysis. Then, attention is turned to testing for a potential naval arms race in Asia. Finally, I present some discussions and offer some conclusions about the findings.

Asian naval growth

A crucial argument for this article is that naval power is on the rise in Asia. An examination of the trends in naval data from Crisher and Souva (2014), along with new naval data collected (discussed below), illustrates the growth of naval power in Asia. In particular, we see that although naval power is on the decline in many parts of the world, it has risen dramatically in Asia.² Figure 1 shows the trends graphically, whereas Table 1 isolates naval growth for specific states.

Figure 1.

Asian versus non-Asian percent change in tonnage since 1960.

Table 1.

Comparing naval growth between Asian and non-Asian powers.

Non-Asian	1960	2015	Percent change
USA	7,628,246	3,558,750	–53
Russia	1,283,973	770,710	–40
UK	1,094,343	256,800	–77
France	530,149	239,210	–55
Argentina	134,974	35,792	–73
Brazil	57,541	109,468	90
Asian	1960	2015	Percent change
China	79,448	716,552	802
Japan	45,682	359,500	687
South Korea	42,352	193,392	357
India	35,392	303,230	757

The graph in Figure 1 shows an overall decline in naval power outside of Asia. Granted, outside of Asia, there is still a total naval tonnage of just over 6.4 m (and starting to rise). Nevertheless, the 6.4 m tons in 2015 is in stark contrast to the 11.6 m tons available in 1960. Conversely, the graph shows the rapid increase in tonnage in Asia since 1960. A region with less than 500,000 tons in 1960 now has over 2 m tons in 2015. The total tonnage would most certainly have continued its rise through 2019 as China, South Korea, and Japan have all continued to launch new ships.

Table 1 highlights these developments by isolating changes in tonnage for a few key actors. Traditional naval powers such as the USA and Russia still maintain sizable navies (the USA in particular). Yet, they have decreased in size since 1960 with the USA seeing a decline of 53%. The British Navy, in particular, has declined by nearly 77%. The Asian naval powers, however, have all grown. China’s navy has grown by 802%, whereas Japan’s navy has grown by 687%. Even South Korea’s navy has grown by 357%, and the commissioning of its newest amphibious assault ship, the 18,000-ton Dokdo, will further bolster these numbers. Although the Asian navies are not on a level with the US Navy, Asia does have navies capable of competing with European navies. The region, in general, is witnessing an overall growth in naval might.

Explaining naval growth

Numerous factors could potentially explain naval growth. The first set of factors emphasizes economic explanations. Simply put, navies are expensive. As such, states seeking to expand their naval might will need the necessary resources to do so. At the most basic level, we should expect states to grow their naval might as their economies grow. Gilady (2018) did find evidence linking GDP to naval growth, but she primarily argued that states pursuing capital ships, such as aircraft carriers, were doing so as vanity projects. However, this is likely an oversimplistic view of naval development.³

Further economic explanations focus on development rather than just economic resources. Markowitz et al. (2019) argue that as a state’s economy develops, its sources of wealth change. In particular, states become less dependent on land rents and shift their attention to overseas economic opportunities. As these states become economically more outward-looking, they develop greater naval capabilities to protect their access to overseas markets. Similarly, there is some evidence that states will develop their power projection capabilities as their levels of international trade increase (Fordham, 2011). As such, a state’s level of trade and its economic development level may have an impact on naval might.

The second set of factors focuses on geostrategic explanations. Competition is at the heart of the international system (Mearsheimer, 2001; Waltz, 1979). Markowitz and Fariss (2018) argue that as a state’s competitive environment increases, it is more likely to respond with expanded military capabilities. In particular, states have a higher likelihood of developing naval might as their competitive environment increases. A vital aspect of the Markowitz and Fariss (2018) argument is that as the distance between rivals decreases, the overall threat environment increases. The emphasis on distance would imply that regional rivals have a greater threat environment. Hence, Asian states in a highly competitive environment should have greater naval power.

Similarly, Mahan (2004) made the argument that a state’s geographic position will play a key role in its naval development. Although Mahan (2004) emphasized the length of a state’s coastline and quality of ports for naval development, other geographical features could also have an impact on this.⁴ In particular, states that are situated near naval chokepoints have a greater incentive to develop naval might. Chokepoints are the lifeblood of a global economy and serve as crucial strategic bargaining chips. Asian states that are situated near naval chokepoints, for example, the Strait of Malacca, should have more considerable naval might.

In sum, there are two primary sets of factors that could be driving naval growth in Asia. One set of factors focuses on economics. An increase in overall economic resources or trade levels could be driving growth. A slightly different economic story is that states developing beyond a land rent economy are more likely to develop naval power. The second set of factors focuses on geostrategic explanations. A state with an increasing threat environment may respond with greater power projection capabilities. Finally, being situated near key strategic chokepoints may encourage states to increase their naval capabilities.

Naval data and research design

The temporal domain for the analysis is 1960–2015 because of the limitations of the economic and naval data.⁵ The spatial domain for the analysis is states in Asia with some naval power (i.e., states identified in the Crisher and Souva (2014) data set). Two different coding schemes are used for selecting whether states are situated in Asia. One scheme comes from Danilovic (2001), the other from the COW project. Table 4 in the Appendix shows the differences between the Danilovic and COW classifcations. Regardless of the coding scheme, the primary findings remain the same.⁶

Dependent variables

Quantifying naval power is a difficult task. Crisher and Souva (2014) have made the task less onerous with the release of their naval data set. However, a limitation of the data set is the failure to include major amphibious platforms (Crisher and Souva, 2014: 609).⁷ Nevertheless, major amphibious platforms could be a way for states to bolster their naval might without incurring expenditure on capital ships. Such platforms allow states to project power through amphibious assaults. They can also serve in a sea denial role by focusing on coastal defense through tasks such as anti-submarine warfare. More advanced amphibious ships could serve in a sea control role. In particular, China has placed a greater emphasis on developing amphibious platforms since the mid-1990s (Cole, 2016: 61).

Data on amphibious platforms were collected to serve as an addendum to the Crisher and Souva (2014) data set and remedy the shortcoming. The coding followed the procedures described in Crisher and Souva (2014). Data for 1960–1994 were coded from Conway’s All the World’s Fighting Ships (Gardiner et al., 1995). For 1995–2015, the data are coded from The Military Balance, which is published by the International Institute for Strategic Studies. The data include information on assault ships, landing helicopter docks, landing platforms helicopter, landing platform docks, dock landing ships, tank landing ships (LSTs), landing ship logistics, helicopter support, command ships, attack transport, attack cargo, and helicopter carriers.⁸ Ship types were coded down to the size of LSTs because they are all classified as amphibious warfare platforms with full data coverage. Other types of amphibious platform exist, for example, tank landing craft, for which the data on ship classes, precise numbers in service, and overall service lives were far too limited to include in the data set. Tonnage for the amphibious ships was then merged with the Crisher and Souva (2014) naval data to create an expanded measure of a state’s Tonnage to serve as the primary dependent variable. Because of the skewed distribution of naval tonnage, the natural log of Tonnage is calculated before its inclusion in the regression models.

Independent variables

One set of factors driving naval development in Asia focuses on economic explanations. Developing naval might takes resources. As such, GDP should have a positive impact on naval development. Here, Ln GDP represents the natural log of a state’s GDP using data from the World Bank (2020). Markowitz et al. (2019) make the argument that states less focused on land rents will support the development of naval might to protect overseas rents. They use energy consumption per capita as their measure of state economic interest. Similarly, I use data from the Correlates of War (COW) National Materials Capabilities data set to create Ln Energy, which represents the natural log of a state’s energy consumption per capita (Singer et al., 1972).

Finally, to capture the potential impact of trade policies on naval development, Trade represents the product of a state’s level of trade openness and its GDP. In other words, Trade is calculated as Trade as a percentage of GDP * GDP. The measure has the advantage of distinguishing between large-volume traders from large economies and large-volume traders from small economies.⁹ The data are taken from the World Bank.

There are also geostrategic explanations for naval development. Markowitz and Fariss (2018) argue that increased competition could lead to states developing naval power. They create a new measure of competition, which produces a single measure of competition that incorporates measures of relative economic power, geographic proximity, and interest compatibility. In the regression analysis, Competition is the Markowitz and Fariss (2018) measure of competition.¹⁰ Additionally, I include Chokepoint as a binary measure of whether a state is situated next to a naval chokepoint.¹¹ Chokepoints are crucial strategic and economic waterways. States situated next to these chokepoints have more significant incentives to invest in naval development.

Finally, I include two additional controls in the regression models. Democracy is a binary measure of democracy using the Polity IV data (Marshall and Jaggers, 2009). Specifically, the measure equals one for states with a combined democracy and autocracy score of higher than six and equals zero otherwise. Additionally, System Tonnage is included as an overall measure of total tonnage in the system (excluding Asia).¹² Including the total tonnage in the system will allow us to capture the potential influence of the shrinking non-Asian navies, for example, that of the USA, on regional naval development.

Model selection

As the dependent variable is continuous, I use a least squares model with robust standard errors. To mitigate autocorrelation concerns, I include a lagged dependent variable with the other independent variables also lagged. The independent variables have a three-year lag. The decision to use a three-year lag structure is discussed in greater depth by Bolks and Stoll (2000: 593–594). In short, a one-year lag is inappropriate because it takes multiple years for the construction and commissioning of warships. For robustness, all models were also run with a five-year lag structure.

Findings for state year models

Table 2 shows the results of the regression models. The various models test the economic and geostrategic factors using different time lags and different definitions of Asia (Danilovic versus COW), and also differentiate between Cold War and post-Cold War naval development.

Table 2.

Results for the determinants of Asian naval power.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
	1960–1989	1990–2015	1960–1989	1990–2015	1960–1989	1990–2015	1960–1989	1990–2015
	3-year lag	3-year lag	3-year lag	3-year lag	5-year lag	5-year lag	5-year lag	5-year lag
	Danalovic	Danalovic	COW	COW	Danalovic	Danalovic	COW	COW
Lagged Ln Tonnage	0.254*** (0.040)	0.120** (0.037)	0.262*** (0.044)	0.113** (0.040)	0.188*** (0.027)	0.107*** (0.028)	0.188*** (0.029)	0.095** (0.030)
Competition	601.940*** (117.337)	−51.623*** (15.509)	592.388*** (113.831)	−20.550 (15.183)	607.553*** (130.353)	−25.315 (13.998)	565.335*** (129.086)	7.147 (14.558)
Ln Energy	0.138*** (0.029)	0.231*** (0.029)	0.141*** (0.019)	0.180*** (0.026)	0.175*** (0.031)	0.251*** (0.024)	0.153*** (0.020)	0.195*** (0.022)
Trade	−0.000*** (0.000)	−0.000* (0.000)	−0.000*** (0.000)	−0.000* (0.000)	−0.000*** (0.000)	−0.000** (0.000)	−0.000*** (0.000)	−0.000** (0.000)
Ln GDP	0.571*** (0.050)	0.777*** (0.042)	0.597*** (0.046)	0.858*** (0.046)	0.597*** (0.046)	0.806*** (0.034)	0.650*** (0.039)	0.890*** (0.037)
Democracy	2.167*** (0.537)	0.010 (0.087)	2.104*** (0.508)	0.054 (0.085)	2.198*** (0.583)	0.033 (0.083)	1.920*** (0.571)	0.078 (0.084)
Chokepoint	−0.059 (0.082)	0.101* (0.049)	−0.078 (0.083)	0.218*** (0.054)	−0.047 (0.084)	0.105* (0.047)	−0.074 (0.084)	0.224*** (0.050)
System Tonnage	−0.000* (0.000)	0.000*** (0.000)	−0.000 (0.000)	0.000*** (0.000)	−0.000* (0.000)	0.000*** (0.000)	−0.000 (0.000)	0.000*** (0.000)
Constant	−0.950 (0.832)	−1.556*** (0.414)	−2.182** (0.765)	−2.789*** (0.410)	−0.645 (0.882)	−1.879*** (0.385)	−2.341** (0.837)	−3.117*** (0.391)
N	257	289	301	341	237	289	277	341

COW: Correlates of War.

p ⩽ 5%, **p ⩽ 1%, ***p ⩽ 0.1% for two-tailed tests with robust standard errors.

From the models, we see two factors consistently have a statistically significant impact on Asian naval development. In particular, the findings strongly support an argument of naval development in Asia driven by economic development.

First, energy consumption has a positive and statistically significant impact on naval development regardless of model selection. As Markowitz et al. (2019) argue, economies that become more outward-looking are more likely to engage in naval development. The finding for energy consumption offers support for this line of argument.

The logarithmic transformation of the energy consumption and naval tonnage variables makes interpretation of the coefficient a little tricky. From Model 1, increasing a state’s energy consumption by 35% will result in a 4% increase in its naval tonnage.¹³ Note that these increases would have a significant impact on relatively less developed states (such as China) rather than developed states (such as the USA) because of the variable’s logarithmic nature. For example, a 32% growth in China’s energy consumption in 1980 represents roughly an increase of 0.27 in energy consumption. In comparison, the same 32% increase for the USA in 1980 would mean a 3.6 increase in energy consumption. In other words, the USA’s energy consumption growth would need to be significantly greater than China’s to realize the 4% growth in naval power. A 3.6 increase in China’s energy consumption would result in roughly a 50% growth in naval might versus the 4% for the USA. As such, we see that the finding is statistically and substantively significant.

Second, we see that GDP has a positive and statistically significant impact on naval development regardless of model selection. Such a finding is unsurprising given the resource-intensive nature of developing and maintaining naval power. In terms of substantive impact, a 12% increase in GDP increases a state’s naval tonnage by 7%.¹⁴ Similar to energy consumption, increasing GDP would have a greater impact on the navies of relatively small economies rather than larger ones. For example, in 1980, US GDP would need to increase by 798,810 (millions) to have the same 12% increase as China (needing an increase of 174,681(millions)). In other words, modest increases in Chinese GDP at this time could have a large impact on the overall size of China’s navy. Again, we see a statistically and substantively significant finding.

With regard to energy consumption and GDP, Model 2 suggests that the impact of these variables becomes stronger in the post-Cold War era. For both variables, we see a larger coefficient in Model 2. Figure 2 shows the relationship graphically. On the left, we see the impact of energy consumption on a state’s naval tonnage as energy consumption increases from roughly its 5th to 95th percentile. We see the positive influence of energy consumption and that it is greater in the post-Cold War era. The same 35% increase in energy consumption now increases a state’s tonnage by 7% (it was 4% during the Cold War).

Figure 2.

Predicted tonnage from models 1 and 2.

Similarly, the graph on the right shows the impact of GDP on a state’s naval tonnage as GDP increases from roughly its 5th to 95th percentile.¹⁵ The difference between the Cold War and the post-Cold War era is not quite as distinct for GDP. Here, the same 12% increase in GDP now increases a state’s tonnage by 9% rather than the 7% we saw in Model 1. In either case, the economic indicators are having a substantively significant impact on Asian naval development.

The third economic measure, Trade, shows a negative and statistically significant relationship with naval development. A negative relationship is surprising as it would be expected that as states trade more, they would develop naval power. However, every one-unit increase in Trade (in millions), decreases a state’s naval tonnage by 0.000056%. In other words, although the relationship is statistically significant, substantively, the relationship is marginal.

The lone geographic variable provides an intriguing result as well. In the post-Cold War era, we see that states situated near a chokepoint are increasing their naval tonnage. However, during the Cold War, this is not a statistically significant relationship. Considering the region hosts four chokepoints, this should not be surprising.

But that the relationship between chokepoints and naval development is time-dependent is worthy of exploration in future research. It could be argued that during the Cold War, the USA was preserving the public good by protecting chokepoints for global trade, but this is reversed in the post-Cold War era.

Another interesting finding from the models is the inconsistent results for global competition. Markowitz and Fariss (2018) argue that states will increase their naval tonnage in response to increasing competition. The findings from the Cold War models show that this is indeed the case because the coefficient for Competition is positive and statistically significant. However, the post-Cold War era paints a far different story. Model 2 shows a negative and statistically significant relationship between competition and naval development. However, the result is not robust for alternative model specifications. In the post-Cold War models, we see that competition is insignificant in Models 4, 6, and 8 according to the different definitions of Asia and changing lag structures. The result does suggest a potential temporal effect for competition. Additionally, as the Markowitz and Fariss (2018) measure could more accurately be thought of as a measure of economic competition, the post-Cold War results suggest a different type of competition is driving naval development. Let us now turn to that possible competition.

A new naval arms race?

From the neorealist perspective, states have two ways of dealing with rivals that are growing militarily. One option is to seek help by way of allies. The logic is simple. If a state is not powerful enough to deter a rival, combining the power of multiple states in an alliance could do the trick. Such balancing is known as external balancing (Mearsheimer, 2001; Waltz, 1979).

Although external balancing is one option, it does leave a state open to abandonment. The anarchic nature of the international system means it is always possible that an ally will fail to meet its obligations at the moment of truth. As such, states can attempt to balance a rising rival by building up their military might. After all, the dogma of self-help characterizes the international system, so the best way to help oneself is through military expansion (Parent and Rosato, 2015: 85). Such balancing is known as internal balancing (Mearsheimer, 2001; Waltz, 1979).

Despite the possibility of internal balancing, examining naval development in the Asian seas is often just another way to view the rivalry between China and the USA. In this sense, China focusing on internal balancing is not difficult to understand. After all, there are not many potential allies lining up to stand side by side with China against the USA.¹⁶ To gain freedom of action within its region, China must develop the necessary naval capabilities (Leon, 2017: 18).

However, thinking about naval power in the Asian seas as primarily a question of the USA versus China is problematic (Beckley, 2017: 80). Although China may have one eye on the US Navy, it will be concerned about its regional rivals – Japan in particular – having the potential for naval growth (Christensen, 2003: 29). After all, should China want to break out of Asian waters, the navies of Japan and India are well positioned to make life difficult for the People’s Liberation Army Navy (PLAN). Because of China’s heavy dependence on energy imports, it is quite vulnerable to the disruption of trade should a regional navy target a chokepoint such as the Strait of Malacca, which links China to the Indian Ocean and oil from the Middle East (Shambaugh, 2013: 291). One of the key ideas from the naval theorist Corbett, is that smaller navies can engage larger navies through sea denial (Corbett, 2004). Hence, China is in the difficult position of playing catch-up with the US Navy while also needing to monitor other navies growing in the region.

Hyperfocusing on the developing USA–China rivalry often leads to the assumption that the other Asian powers will primarily employ external balancing. Such a myopic view would lead to the expectation that Japan and South Korea would depend mostly on the protective blanket of the US Navy. However, there is a growing concern that the USA might not be the dependable partner it once was. Concerns with abandonment in relation to Japan began in the early 1990s (Dreyer, 2016: 296), and subsequent US administrations have done little to ameliorate the problem (Inoguchi, 2011: 238). South Korea, too, has similar concerns about the USA (Kim, 2018).

In terms of external balancing, Levy and Thompson (2010) note that naval hegemons will be less likely to encourage such counterbalancing coalitions. However, Blagden et al. (2011: 193) argue that a growing naval power could trigger both external and internal balancing.¹⁷ Hence, as doubts about the willingness of the USA to risk expensive naval assets for its Asian allies continue to grow, Japan and South Korea are more likely to look toward internal balancing. Prudence and the dictates of self-help suggest this is a proper course of action for the two countries.

The offense–defense balance would argue that Japan and South Korea would be better off working on anti-access/area-denial (A2/AD) for territorial protection than countering expanding Chinese naval development (Beckley, 2017: 81). A2/AD capabilities would focus on minimizing the ability of Chinese naval capabilities from directly impacting Japanese or South Korean homelands. Yet, these countries are economically vulnerable as well. A hegemonic China can simultaneously shut down multiple chokepoints in the region and devastate South Korea’s and Japan’s economies.¹⁸ Hence, these two countries need to develop some naval might that would give China pause for thought before employing such tactics. In other words, there is a difference between protecting only territorial interests and protecting economic interests that are extra-regional.

We should also remember that although Japan and South Korea are likely to engage in internal balancing against China, they are likely to balance against each another as well.

Japan and South Korea have long-term animosities stemming from Japan’s past occupation of Korea. These animosities still strain relations today (Easley and Park, 2018). Because of these long-standing issues and their proximity to each another, Japan and South Korea are likely to respond to the expansion of the other’s navy as well.

India seems to be a case for which we would also expect to see internal balancing. Concerns about China have long dominated Indian foreign policy (Pant and Rej, 2018). Expanding Chinese interests into the Indian Ocean with the “String of Pearls” strategy is particularly disconcerting. China’s String of Pearl strategy sets up a series of military and commercial facilities to link China across the Indian Ocean all the way to the Horn of Africa. Because India is also surrounded by crucial economic shipping routes, we would expect it to try and balance internally as China grows in might.

Naval arms race analysis

Model selection

A naval arms race implies that for some states, their decisions with regard to naval development will be dependent on the naval development of some rival states. To capture this potential relationship, I utilize a seemingly unrelated regression (SUR) model (Zellner, 1962).¹⁹ SUR models use multiple linear equations and test whether the error terms of the model are correlated. For these models, I model the tonnage for each state and measure total tonnage for their specific rivals. The SUR models allow us to see if the decisions of the main Asian naval powers are dependent upon their rivals’ actions. If the models are unrelated, we are simply left with a series of ordinary least squares (OLS) estimates (Bolks and Stoll, 2000).

The SUR models will focus on the primary naval powers in the region, in particular, China, Japan, India, and South Korea, as shown in Table 1. The variable Rival Tonnage represents the tonnage for each state’s primary rivals.

Japan and India serve as the rivals for China. China and South Korea serve as the rivals for Japan. China and Japan serve as the rivals for India. Finally, China and Japan serve as the rivals for South Korea.²⁰

A final point in relation to the SUR models that must be considered: because we are now modeling the tonnage for individual states, the number of observations per model will greatly decrease.

As such, the SUR models cannot include all of the variables seen in Table 2. Rather, the SUR models will include Lagged Tonnage, Energy, Rival Tonnage, and Trade.²¹

SUR findings

Table 3 shows the results of the SUR models. The first findings of note are Breusch–Pagan tests (chi-square). Remember that if the models are independent, the SUR model reduces to a series of OLS estimates. The Breusch–Pagan test considers the independence of the dependent variables with a null hypothesis of no relationship. For each of the SUR models, the chi-square statistic is of sufficient value to reject the null hypothesis of independence at the 90% level. In other words, the SUR models provide evidence that the naval development of the Asian powers is related.

Table 3.

Results for the seemingly unrelated regression models.

	Model 9	Model 10	Model 11	Model 12
	1960–1989	1990–2015	1960–1989	1990–2015
	3-year lag	3-year lag	5-year lag	5-year lag
China
Lagged Tonnage	0.384** (0.123)	0.631*** (0.137)	0.099 (0.103)	0.172 (0.167)
Trade	−0.000*** (0.000)	0.000* (0.000)	−0.000*** (0.000)	−0.000 (0.000)
Energy	78630.298*** (15211.163)	−1.01e+04 (42068.958)	89456.353*** (8950.370)	1.05e+05* (44641.251)
Rival Tonnage	0.348*** (0.086)	−0.087 (0.081)	0.596*** (0.068)	1.147*** (0.281)
Constant	−1.38e+04 (7476.896)	1.63e+05* (63538.240)	−1.70e+04** (5515.035)	−3.16e+05** (1.06e+05)
Japan
Lagged Tonnage	1.002*** (0.138)	0.579*** (0.130)	0.751*** (0.193)	0.468*** (0.138)
Trade	−0.000 (0.000)	−0.000 (0.000)	0.000 (0.000)	−0.000*** (0.000)
Energy	−1150.148 (2363.758)	35189.714*** (6851.099)	−3786.617 (2527.105)	28036.009*** (6909.500)
Rival Tonnage	0.237* (0.111)	−0.002 (0.020)	0.341** (0.123)	0.079* (0.034)
Constant	−4726.546 (7535.822)	−2.53e+04 (24825.551)	3901.497 (8877.345)	50293.149* (21715.152)
India
Lagged Tonnage	0.521* (0.220)	0.508 (0.276)	−0.089 (0.432)	−0.144 (0.227)
Trade	−0.000* (0.000)	−0.000 (0.000)	−0.000* (0.000)	0.000 (0.000)
Energy	−1.11e+05* (54986.077)	1.45e+05** (53955.487)	−1.79e+05** (55798.598)	1.23e+05 (87936.679)
Rival Tonnage	0.468*** (0.137)	−0.041* (0.017)	0.766*** (0.165)	0.080 (0.066)
Constant	−3572.044 (5939.288)	33503.510 (33832.832)	3977.298 (10442.586)	88398.307*** (26456.988)
South Korea
Lagged Tonnage	0.203 (0.222)	−0.061 (0.175)	0.201 (0.254)	−0.735*** (0.161)
Trade	−0.000 (0.000)	−0.000 (0.000)	−0.000 (0.000)	−0.000* (0.000)
Energy	9301.400 (5462.747)	23232.605*** (4314.114)	9062.367 (6498.919)	22984.907*** (4447.401)
Rival Tonnage	0.088 (0.047)	0.000 (0.014)	0.150* (0.063)	0.161* (0.074)
Constant	25651.631** (8451.578)	28016.509** (8968.842)	20297.712 (10637.068)	25409.422 (18476.882)
N	27	26	25	26
χ2	19.8 (p ¡ 0.01)	9.9 (p ¿ 0.1)	36.9 (p ¡ 0.01)	(p ¡ 0.1)

p ⩽ 5%, **p ⩽ 1%, ***p ⩽ 0.1%.

The primary point of interest in these models is the impact of rival tonnage on naval development. The models point to Japan and China as the countries with the most consistent impact of rival tonnage on naval tonnage. Except for Model 10, there is evidence of Japan and China developing in response to regional rivals. If we focus on Models 11 and 12, it is interesting to note the change in rival tonnage coefficients for Japan and China. The effect of increasing rival tonnage has a more significant impact on China in the post-Cold War era, whereas we see the opposite for Japan. In other words, China is accelerating naval development, whereas, seemingly, Japan is slowing it down. As discussed below, the fact that Japan is slowing down naval development makes sense as it reaches the limits of what its resources will allow it to develop.

There are two forces likely driving Japan’s increase in naval power. One factor is the fear of US abandonment, whereas the other is the USA encouraging Japan to take on more responsibility for security in the region (Christensen, 2003; Dreyer, 2016; Easley and Park, 2018). Related to the second factor, Japan taking on more responsibility in the region could also be a way for the country to balance its military relationship with the USA. In effect, the Japanese Maritime Self-Defense Force (JMSDF) could potentially respond to the growth of the South Korean Navy, and this is a reminder that the regional rivalries in the region exist outside of China’s presence. The fact that China seems to be responding more to regional rivals post-Cold War is interesting. In 1996, President Clinton sent not one, but two aircraft carriers through the Taiwan Strait, and it could be imagined that this occurrence would have placed China’s focus primarily on the USA. In point of fact, the action spurred the Chinese into growing their navy to gain more independence in their region (Cole, 2016; Shambaugh, 2013).

However, China needs to keep an eye on the JMSDF, and other regional powers as well. Cole (2016: 197) reminds us that “Japan, however, is the historical enemy, very much a maritime nation, and, hence, a primary concern to PLAN planners and strategists.” Additionally, an article by a Chinese naval strategist in 1990 specifically discussed concerns in relation to a growing JMSDF at the behest of the USA and more robust ties between the Japanese and South Korean navies (Dreyer, 2016: 295). As such, finding evidence that the Chinese Navy is growing in response to the growth in the Japanese and Indian navies is less surprising.

The models in Table 3 also show evidence that South Korea is reacting to Japan and China. The lone exception is Model 10, which uses a three-year lag for the post-Cold War era.²² South Korea finds itself in a similar position to Japan. As confidence in the US commitment to the region continues to decline, South Korea faces a less than desirable geopolitical position. Additionally, South Korea may also desire more balance in its military relations with the USA. As such, it is not surprising that its naval might is growing along with that of China and Japan. The country’s desire for a blue water navy anchored by the Dokdo class of amphibious assault ships drives its pursuit of greater security independence.

Perhaps the most interesting finding from SUR models concerns India. India does not seem to respond to the naval development of China and Japan post-Cold War. Again, this seems odd, because we would expect India to counter China’s naval growth, particularly in the post-Cold War era. The null finding, and negative finding of Model 10, could highlight a limit to Indian resources because the country already operates more than one capital ship. Perhaps India has reached a point of no longer being able to expand its naval might while continuing to battle internal issues. It also has a problem in that it borders two states with long-standing territorial issues – China and Pakistan. Pant and Rej (2018: 49) note that India is placing greater emphasis on land threats than maritime ones. The finding reminds us that states do not have unlimited resources to devote to land and naval power. Here, India has a finite amount of resources to dedicate towards their land and naval forces and cannot afford to allow the navy to expand too much.

Discussion and conclusions

The SUR results suggest we are witnessing a developing Asian naval arms race in the post-Cold War era. Nevertheless, there are reasons to temper the implications of these findings. One reason is that the SUR models do not provide definitive evidence of a causal relationship between the Asian powers’ naval development. For example, if China’s primary focus is on closing the capability gap with the USA, the PLAN would grow even if the JMSDF were growing. Similarly, if Japan is primarily concerned with developing a minimal self-defense capability independent of the USA, the JMSDF would grow even if the Chinese Navy were growing. Hence, although the results are certainly suggestive, more work needs to be done.

A second reason is the data limitations, which prevent the analysis from going beyond 2015. Updated data could help to show whether the naval development of the post-Cold War era is indeed evidence of an Asian naval race. The latest major ships commissioned by China, Japan, and South Korea are not included in this analysis. Updating the available naval data, along with economic indicators such as energy consumption, could provide more clarity as to whether there is a naval arms race.

Additionally, resource constraints with regard to Japan, South Korea, and India ensure they can only grow so much. Much like the Germans before World War I who could not build indefinitely to match the British, the naval might of the Asian powers will level out. Perhaps we are starting to see this already with India. There was a race to develop a certain defense against a growing China, but there was a point at which this stopped. Similarly, budgetary constraints that may be holding Japan back in relation to additional naval development could lead to a plateau here as well (Heginbotham and Samuels, 2018). If Japan were beginning to reach its limits, this would explain the smaller coefficient for Rival Tonnage in Model 12. However, both Japan and South Korea have commissioned major warships in the last few years. Japan commissioned the 27,000-ton helicopter carrier Izumo (undergoing a conversion to a light aircraft carrier), and South Korea commissioned the 19,500-ton amphibious assault ship Dokdo. The launching of two major surface warships by each state suggests that further growth is possible should they desire it. Indeed, South Korea has another Dokdo-class assault ship (Marado) undergoing sea trials.

Limited resources, in particular, could help us understand the lack of evidence for the Indian navy growing in response to regional rivals in the post-Cold War era. India may feel it has sufficient naval forces for its current needs with an aircraft carrier in service and a new nuclear-powered ballistic missile submarine. Because Japan and South Korea have lacked major capital ships and an independent nuclear deterrent, they have felt the need to prioritize more significant naval development in the post-Cold War era.

Overall, the results of the article suggest that a combination of economic and geostrategic factors is driving naval growth in Asia. At the regional level, economic development provides Asian states with the opportunity to invest in their naval might. However, geopolitics is providing the willingness to develop naval power. Surprisingly, economic competition does not seem to be driving modern naval development. Instead, we see evidence that the economic development in Asia has allowed Asian navies to grow in response to the naval growth of their regional rivals.

One future avenue of exploration with regard to the growth of naval might in the region lies in domestic politics. Decisions in relation to the development of military might do not occur in a vacuum, and states must decide how to allocate their resources. For example, Fordham (2019) explores how domestic political battles shaped the formation of the US battle fleet at the end of the 19th century. Buzan and Herring (1998) offer another potential framework for understanding how domestic politics and structures can have an impact on the composition of a state’s military force. A more in-depth examination of the Asian naval powers would augment the primary findings of this article.

Regardless of the primary explanation for Asian naval growth, we can certainly see a dramatic increase in naval capabilities in the region. What this means for the stability of the region in the future remains an open question. With few exceptions, the links between naval growth and conflict is an understudied topic.²³ Future work on the topic would do well to keep in mind the regional implications of such research. The unique combination of naval growth, rivalry, and maritime disputes may provide the perfect recipe for future conflict in the Asian region.

Footnotes

Appendix

Table 4.

Danilovic and COW lists of Asian states.

States in both lists	States in COW only
Afghanistan	Turkmenistan
Tajikistan	Uzbekistan
Kyrgyz Republic	East Timor
Kazakhstan	Australia
China	Papua New Guinea
Mongolia	New Zealand
Taiwan	Vanuatu
Korea, North	Solomon Islands
Korea, South	Kiribati
Japan	Tuvalu
India	Fiji
Bhutan	Tonga
Pakistan	Nauru
Bangladesh	Marshall Islands
Myanmar	Palau
Sri Lanka	Federated States of Micronesia
Maldive Islands	Samoa
Nepal
Thailand
Cambodia
Laos
Vietnam
Vietnam, Republic of
Malaysia
Singapore
Brunei
Philippines
Indonesia

COW: Correlates of War.

Acknowledgements

I would like to thank Mark Souva and Michelle Williams for their helpful suggestions. I would also like to thank John Link and Robert Lee for their hard work coding the data on amphibious warships. Finally, I am grateful to the editors and reviewers at the Journal of Asian and African Studies for providing valuable feedback and suggestions for this project.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Brian Benjamin Crisher

Notes

Author biography

Brian Benjamin Crisher is an assistant professor at the University of West Florida. He has written on military capabilities – specifically naval power – and their impact on war and politics.

References

Beckley

(2017) The emerging military balance in east Asia: How China’s neighbors can check Chinese naval expansion. International Security 42(2): 78–119.

Blagden

Levy

Thompson

(2011) Sea powers, continental powers, and balancing theory. International Security 36(2): 190–202.

Bolks

Stoll

(2000) The arms acquisition process: The effect of internal and external constraints on arms race dynamics. Journal of Conflict Resolution 44(5): 580–603.

Buzan

Herring

(1998) The Arms Dynamics in World Politics. Boulder, CO: Lynne Rienner.

Chang Liao

N-C

(2018) Winds of change: Assessing China’s assertive turn in foreign policy. Journal of Asian and African Studies 53(6): 880–895.

Christensen

(2003) China, the U.S.–Japan alliance, and the security dilemma in East Asia. In: Ikenberry

Mastanduno

(eds) International Relations Theory and the Asia-Pacific. New York: Columbia University Press, 25–56.

Cole

(2016) China’s Quest for Great Power: Ships, Oil, and Foreign Policy. Annapolis, MD: Naval Institute Press.

Corbett

(2004) Principles of Maritime Strategy. New York: Dover Publications.

Crisher

(2017) Naval power, endogeneity, and long-distance disputes. Research & Politics 4(1): 1–6.

10.

Crisher

Souva

(2014) Power at sea: A naval power dataset, 1865–2011. International Interactions 40(4): 602–629.

11.

Danilovic

(2001) The sources of threat credibility in extended deterrence. Journal of Conflict Resolution 45(3): 341–369.

12.

Dreyer

(2016) Middle Kingdom and Empire of the Rising Sun. New York: Oxford University Press.

13.

Easley

L-E

Park

(2018) South Korea’s mismatched diplomacy in Asia: Middle power identity, interests, and foreign policy. International Politics 55(2): 242–263.

14.

Fordham

(2011) Who wants to be a major power? Explaining the expansion of foreign policy ambition. Journal of Peace Research 48(5): 587–603.

15.

Fordham

(2019) The domestic politics of world power: Explaining debates over the United States battleship fleet, 1890–91. International Organization 73(2): 435–468.

16.

Gardiner

Chumbley

Budzbon

(1995) Conway’s All the World’s Fighting Ships, 1947–1995. Annapolis, MD: Naval Institute Press.

17.

Gilady

(2018) The Price of Prestige. Chicago, IL: University of Chicago Press.

18.

Heginbotham

Samuels

(2018) Active denial: Redesigning Japan’s response to China’s military challenge. International Security 42(4): 128–169.

19.

Horowitz

(2010) The Diffusion of Military Power: Causes and Consequences for International Politics. Princeton, NJ: Princeton University Press.

20.

Inoguchi

(2011) Japanese ideas of Asian regionalism. Japanese Journal of Political Science 12(2): 233–249.

21.

Kim

(2018) Avoiding being a crushed Prawn and becoming a Dolphin swimming between the two fighting Whales? South Korea’s strategic choice in the face of the intensifying Sino–US competition. Journal of Asian and African Studies 53: 612–628.

22.

Leon

DPY

(2017) Economic interdependence and international conflict: Situating China’s economic and military rise. Asian Politics & Policy 9(1): 9–30.

23.

Levy

Thompson

(2010) Balancing on land and at sea: Do states ally against the leading global power? International Security 35(1): 7–43.

24.

Mahan

(2004) The Influence of Sea Power upon History, 1660–1783. New York: Barnes and Noble.

25.

Markowitz

Fariss

(2018) Power, proximity, and democracy: Geopolitical competition in the international system. Journal of Peace Research 55(1): 78–93.

26.

Markowitz

Fariss

McMahon

(2019) Producing goods and projecting power: How what you make influences what you take. Journal of Conflict Resolution 63(6): 1368–1402.

27.

Marshall

Jaggers

(2009) Polity IV Project: Political Regime Characteristics and Transitions, 1800–2007. Center for Systemic Peace. Available at: http://www.systemicpeace.org/inscrdata.html (accessed May 2020).

28.

Mearsheimer

(2001) The Tragedy of Great Power Politics. New York: WW Norton & Company.

29.

Pant

Rej

(2018) Is India ready for the Indo-Pacific? Washington Quarterly 41(2): 47–61.

30.

Pape

(1993) Why Japan surrendered. International Security 18(2): 154–201.

31.

Parent

Rosato

(2015) Balancing in Neorealism. International Security 40(2): 51–86.

32.

Shambaugh

(2013) China Goes Global: The Partial Power. New York: Oxford University Press.

33.

Singer

Bremmer

Stuckey

(1972) Capability distribution, uncertainty, and major power war, 1820–1965. In: Russett

(ed.) Peace, War, and Numbers. Beverly Hills, CA: SAGE Publications, 19–48.

34.

Stavridis

(2017) Sea Power: The History and Geopolitics of the World’s Oceans. New York: Penguin Press.

35.

Stoll

(2017) To arms, to arms: What do we know about arms races? In: Oxford Research Encyclopedia of Politics. Oxford University Press.

36.

Waltz

(1979) Theory of International Politics. New York: McGraw-Hill.

37.

World Bank (2020) World development indicators: Trade as a percentage of GDP. Available at: https://tcdata360.worldbank.org/indicators/NE.TRD.GNFS.ZS (accessed May 2020).

38.

Zellner

(1962) An efficient method of estimating seemingly unrelated regressions and tests for aggregation bias. Journal of the American Statistical Association 57(298): 348–368.