2. What is the Yoshida Doctrine? Its origins and literature review

The Yoshida Doctrine refers to an approach to Japanese foreign policy initiated by Prime Minister Shigeru Yoshida in the 1950s and established by subsequent politicians, particularly Hayato Ikeda and Eisaku Satō, who succeeded Yoshida in the 1960s. According to the widely held view, the Yoshida Doctrine is based on three main pillars: Japanese security through American alliance, minimized military armaments and expenditures, and an emphasis on economic growth and recovery (Nishihara, Reference Nishihara, Bae-ho and Tadashi1978: 151–155; Nagai, Reference Nagai1985 [2016]: 77–89; Pyle, Reference Pyle1987, Reference Pyle1992: 243–270; Kōsaka, Reference Kōsaka1989: 299; Nakanishi, Reference Nakanishi2003; Soeya, Reference Soeya2008; Sugita, Reference Sugita2016). These pillars are related not concurrently but causally. The US–Japan alliance initially leads to minimal armaments by Japan, which promotes economic recovery and realizes economic growth. In addition to (and closely related to) these three main pillars, the following may be included as the remarkable features of the Yoshida Doctrine: expansion into overseas markets, particularly the Asian market and avoidance of involvement in international political disputes (see Figure 1).

Figure 1. Three main pillars and other two parameters of the ‘Yoshida Doctrine’ or ‘Yoshida Line’.

Reduced military spending in Japan (about 1% of its GDP) was globally unique during the twenty-first century (Figure 2).Footnote ¹ Its accelerated rate of economic growth through expansion into overseas markets was similarly remarkable and has been the subject of international attention, leading several scholars to infer a correlation of the two prominent features of post-war Japan. The foundation of the Yoshida Doctrine is that the Yoshida administration initiated these causal relations after controlling the post-war government.

Figure 2. Percentage of GDP spent on the military by some countries (OECD consists of original members). Source: SIPRI military expenditure database.Footnote ¹⁴

According to a consensus of Japanese foreign policy researchers, the concept of a Yoshida Doctrine or Yoshida Line originated from essays written in the mid-1960s by Masataka Kōsaka, an associate professor at Kyoto University (Nakanishi, Reference Nakanishi2003; Soeya, Reference Soeya2008). A 1964 journal article presented a positive reevaluation of Shigeru Yoshida, who remained unpopular among the public following his administration (Kōsaka, Reference Kōsaka1968: 56, 67, 237–238). Kōsaka based his position on how Yoshida had rebuked intense demands for Japan's rearmament by the USA and focused instead on economic recovery, referring to the Dulles–Yoshida Talks conducted in early-1951, before Japan regained autonomy, as well as the Ikeda–Robertson TalksFootnote ² held in October 1953. In both US–Japan negotiations, Yoshida refused American demands for Japan's rearmament. However, Kōsaka did not use the term ‘Yoshida Doctrine.’ The term was introduced in the late-1970s by political scientists Masashi Nishihara, Yonosuke Nagai, and Kenneth Pyle and used explicitly into the 1980s.

The term was initially used by Nishihara in response to criticism that Japan lacked a ‘diplomatic strategy,’ when he claimed that Japan possessed a ‘strategy originally developed by Yoshida and inherited by successive leaders of the ruling party’ ‘even if the foreign ministry does not wish to acknowledge’ (Nishihara, Reference Nishihara, Bae-ho and Tadashi1978: 151–154). Nishihara said that this strategy, which included expanding the overseas (especially Asian) market, avoiding more than minimum defense spending, and non-involvement in international political disputes, should be called the ‘Yoshida Doctrine’ because it was formulated by Yoshida and continued by his political successors. Assertion of a Japanese foreign policy doctrine was established in 1978, directly following the completion of the ‘Fukuda Doctrine’ that was announced by the Takeo Fukuda administration during meetings with members of the Association of Southeast Asian Nations. Nishihara was likely influenced heavily by the Fukuda Doctrine – the first ‘doctrine’ in Japanese foreign policy – when establishing the term ‘Yoshida Doctrine.’

Afterward, calls for Japan's rearmament increased by those in America who claimed that Japan was getting a ‘free ride’ by achieving economic growth without having to spend money on the military. For example, in 1981, Representative Clement Zablocki issued a concurrent resolution during the 94th Congress calling for ‘Japan's defense expenditures should be at least one percent of its gross national product,’ based on the premise that given its clear economic strength, Japan's contribution to national defense spending is plainly lower than that of other countries of similar stature. In addition, a concurrent resolution proposed in the same year by Representative Stephen Neal appealed to the President to negotiate an agreement that would require Japan to reimburse 2% of its GNP to the USA as a security guarantee tax. In response, there were also calls from some Japanese diplomatic officials referred to as ‘Military Realists’ or ‘Japanese Gaullists’ to expand the military (Mochizuki, Reference Mochizuki1983–1984). Political scientist Yonosuke Nagai, concerned about potential controversy, praised the strategic nature of the Yoshida Doctrine in a paper he presented in 1984. For him, the Yoshida Doctrine referred to the ‘conservative mainstream approach to foreign relations’ based on ‘defensive minimalism’ (1985 [2016]: 78, 258–259). American historian Kenneth Pyle advanced a proposition similar to that of Nagai's (Reference Nagai1985 [2016]: 73; Pyle, Reference Pyle1987). In both cases, the basis for the so-called ‘Yoshida Doctrine’ was attributed to Yoshida's rejection of the demands for rearmament in the Dulles–Yoshida Talks. They claimed that the ‘Yoshida Doctrine’ began to take form during these talks.

Foreign policy based on the Yoshida Doctrine did not prosper during the time of Yoshida's administration or directly afterward; rather, the process began 6 years later, according to the consensus among Japanese researchers (Kōsaka, Reference Kōsaka1968: 258; Iokibe, Reference Iokibe1989: 35; Hatano and Satō, Reference Hatano and Satō2004; Nakajima, Reference Nakajima2006). The Yoshida Doctrine was established during the administrations of Hayato Ikeda (1960–1964) and Eisaku Satō (1964–1972), who are known as the ‘honor students at the Yoshida school.’ It is widely accepted that the ‘Yoshida Doctrine’ was thus inherited in Japan's foreign policy during the 1960s, after the administrations of Ichiro Hatoyama (1955–1956), Tanzan Ishibashi (1956–1957), and Nobusuke Kishi (1957–1960), who succeeded Yoshida and considered themselves political enemies, rejected his political philosophy. Additionally, there is also consensus that Japan was guided by the Yoshida Doctrine until revisionists Junichiro Koizumi and Abe Shinzo came to power (Chai, Reference Chai1997: 389–412; Pyle, Reference Pyle2007; Samuels, Reference Samuels2007; Izumikawa, Reference Izumikawa2010; Dobson, Reference Dobson2017; Soeya, Reference Soeya2017; Kallender and Hughes, Reference Kallender and Hughes2019). Although there is disagreement regarding the actual level of Japanese military power (see Lind, Reference Lind2004), Japan managed to limit defense spending to within 1% of its GDP. Unlike other countries, Japan was pursuing economic centrism and, in this sense, was called realist (Heginbotham and Samuels, Reference Heginbotham and Samuels1998; Green, Reference Green2003). Hence, the Yoshida Doctrine is believed to have been the guiding force in Japan's foreign policy for more than 40 years.

3. Three reasons why it came to be called the ‘Yoshida’ Doctrine or Line

Why was Japanese diplomacy in the latter half of the twentieth century named ‘Yoshida’ Doctrine or Line? Frst, this can be attributed to the existence of some foreign policies decided by the Yoshida administration. This can be further divided into (1) foreign policies that continued after Yoshida, and (2) policy decisions that were limited to the Yoshida cabinet. Second, we can point out (3) some policies match Yoshida's image that emphasizes economic over military power, although the Yoshida cabinet did not actually do them.

The first point seems irrefutable. It was Shigeru Yoshida who signed the US–Japan Security Treaty and gave permission to the US military to stay in Japan. Those policies have continued to this day. As mentioned above, proponents of the ‘Yoshida Doctrine’ emphasize the causal relations of forming an alliance with the USA to curb arms race and achieve economic recovery and growth, that is, the above-mentioned three pillars (see Figure 1). Furthermore, such causal relations were drawn by Yoshida himself had drawn. The second point is regarding the ‘historical fact’ that Yoshida refused when the USA demanded large-scale rearmament. Proponents of the Yoshida Doctrine Kōsaka, Nagai, and Pyle cite the 1951 Dulles–Yoshida meeting as evidence for their arguments. Japan, at that time, was by no means unarmed. However, the refusal of a full-scale rearming request from the USA at the 1951 Dulles–Yoshida Talks and the 1953 Ikeda–Robertson Talks is a symbolically important ‘historical fact’ for Yoshida Doctrine proponents. The third point discusses expanding the overseas (especially Asian) market and non-involvement in international political disputes, which are two features added to the three main pillars. Yoshida emphasized economic over military power and this policy was in line with the characteristics of Japanese diplomacy that followed. Therefore, it was given the proper noun ‘Yoshida.’

The above three reasons, however, have major problems. The problem with the first point is that Japan's foreign policy, as defined by the Yoshida Doctrine, remains intact. The Japan–US alliance has not been revised since 1961, and Japan's military spending remains within the range of approximately 1% of its GDP. Its close alignment with the USA and expansion into overseas markets remains unchanged. Moreover, Japan's economic growth halted in the 1990s and has not changed since the 2000s. Due to evidence of recovery in the Japanese economy since the second Abe administration, it has become impossible to explain the changes in Japan's security policy in this century by arguing for the abandonment of the Yoshida Doctrine. Moreover, the problem with the first point is that the politician's policy is believed to have succeeded if the politician's intentions and actual results match. To assess whether a policy produces appropriate results as intended by the planners, it is necessary to assess it from various angles using different methods. Even short-term policy effects, such as increased minimum wage and changes in employment rate, require a careful assessment. However, proponents of the Yoshida Doctrine do not necessarily pursue the causal mechanism in assessing the long-term policy effects such as the relationship between minimum military spending and economic growth. Yoshida's sketch may indeed have been actualized more than 10 years later, but the process may have been different from what Yoshida had in mind. Yoshida Doctrine scholars (or Japanese diplomatic researchers) have overlooked such a possibility. In this paper, I will examine the causal relationships illustrated in the three main pillars, suggesting the likelihood of a spurious correlation. Quantitative analysis of international datasets was used to reject the idea that Japan was able to maintain minimal armaments because of its alliance with the USA and the stationing of US troops in Japan. Moreover, one ultimate argument is made by presenting recent bodies of research that analyze the relationship between national military expenditures and economic growth to establish that the premise accepted and typically unquestioned by most researchers of Japanese foreign policy – that light armament led to economic growth – is nothing more than the result of a subjective and biased interpretation rather than the result of empirical analysis.

The problem with the second reason is that ‘historical facts’ of the 1951 Dulles–Yoshida Talks and the 1953 Ikeda–Robertson Talks were reconsidered by diplomatic historians in the 1990s. Such academic achievement has been called the ‘Yoshida Doctrine revisionism.’ Since these two cases are important as they have symbolic meaning to the Yoshida Doctrine, the revision of its historical facts will cause great damage to the discourse of the doctrine. In the subsequent section, I will show this ‘Yoshida Doctrine revisionism.’

Finally, the problem with the third reason is, of course, that Yoshida's achievements (responsibility) attribute to what he has not done. For example, the economic expansion into Southeast Asia began in the early-1960s at the earliest. Nishihara (Reference Nishihara, Bae-ho and Tadashi1978: 152) cites expanding external trade into Southeast Asia through reparation payments as one of the characteristics of the Yoshida Doctrine, but only Burma (now Myanmar) has concluded the reparation agreement under the Yoshida administration. The Philippines and Indonesia, which later became the more important trading partners, were slow in negotiations during the Yoshida administration (Nishihara, Reference Nishihara1977; Yoshikawa, Reference Yoshikawa1991). Therefore, it is quite unreasonable to attribute all of Japan's economic expansion into overseas markets to the Yoshida administration. However, since Yoshida himself aimed at economic recovery and growth as a main priority, Japan's later economic growth under his students Ikeda and Satō cabinets can be called the ‘Yoshida Doctrine’ to some extent. This serious problem contains another feature, namely non-involvement in international political disputes because recent studies have revealed that the Yoshida administration was more actively involved in international conflicts compared to the subsequent administrations. In this study, I will take up two of these cases – Yoshida's proactive policy toward China and Japan's involvement in the Korean War. In particular, the latter case can be called the least likely case in the sense that the Yoshida administration itself was involved in an international dispute that the ‘Yoshida Doctrine’ did not anticipate, and could be strong counter-evidence to the discourse of the ‘Yoshida Doctrine.’

This study, therefore, reconsidered and deconstructed the ‘Yoshida Doctrine’ or ‘Yoshida Line,’ referencing multiple related arguments and evidence. Before doing this, the following section shows how the Yoshida Doctrine differs from other foreign policy doctrines and does not merit being called a doctrine.

4. Reconsidering foreign policy doctrines: the unusual nature of the Yoshida Doctrine

Doctrines in the realm of foreign policy are not unanimously defined, meaning that anyone can refer to the foreign policy of any administration as a ‘doctrine.’ For example, Edström (Reference Edström1999: 3) defines Japanese foreign policy doctrines as ‘the set of general and/or specific notions and ideas, openly accounted for by Japan's foreign policy decision-makers, concerning international relations and the international system, the role of Japan in that system, its relationship to other actors in that system, as well as its long-term goals.’ This is a loose definition, and when using it, one could say that every administration in history had a foreign policy doctrine. In practical terms, Edström (Reference Edström1999) analyzed the administrative and general policy speech (shisei hoshin enzetsu and shoshin hyōmei enzetsu) regarding policy from 1952 to 1993, attributing foreign policy doctrines to every administration during this period.

However, by this definition, the ‘Yoshida Doctrine,’ as argued in prior research, would not be considered a foreign policy doctrine because there are no governmental policies or policy speeches that specifically articulate the three pillars (and additional three features) of the Yoshida administration. Although the Yoshida administration pursued reconstruction in its beginning, just as Tetsu Katayama and Hitoshi Ashida had before him in the late-1940s,Footnote ³ in the latter part of his term, from mid-1953 onward, the term disappeared from usage (Edström, Reference Edström1999: 13). Thus, if it was an analytical concept developed by researchers more than 20 years later, the doctrine clearly does not conform to Edström's definition. Therefore, this section considers foreign policy doctrines that have been inductively collected and comparatively examines the features of each in relation to the Yoshida Doctrine. The country that has produced the most foreign policy doctrines is naturally the superpower, the USA. Accordingly, the substance of American foreign policy doctrines is categorized to reveal the unusual nature of the Yoshida Doctrine.

Since the Monroe Doctrine, which was a declaration by the fifth US President, James Monroe, declaring that the nation would refuse colonization and interference by Europe in the Americas and would refrain from interfering in European affairs, American leaders have articulated numerous doctrines (Table 1). However, for 100 years following the 1823 Monroe Doctrine, there were no statements of belief that could be referred to as a doctrine, or at least none that survived. For example, the corollary, proclaimed by Theodore Roosevelt in 1904, which regards foreign policy is understood as a supplement to the Monroe Doctrine. Monroe was the only person to have his name attached to foreign policy doctrine until the Stimson Doctrine in 1932, and during this period in American history, the Monroe Doctrine served as a plan for all American foreign policies.Footnote ⁴

Table 1. US foreign policy doctrines

While these foreign policy doctrines were missing a standard format, several would be declared following World War II, beginning with the Truman Doctrine, which sought to contain communism and which is recognized as one of the most important doctrines in US foreign policy. The Nixon Doctrine, which was mentioned at a press conference in Guam and sought self-help efforts from allies in Asia, was also a predominant US policy.Footnote ⁵ These historically established US foreign policy doctrines exhibit a common characteristic. That is, they were clearly articulated at a particular place and time by a political leader (Brands, Reference Brands2006: 2).

Political declarations of a government must have contemporaneous domestic support, and because these policies were established with an expectation that foreign actors would respond with policies of their own, they required clear and succinct statements. For example, the Monroe and Carter Doctrines were announced in annual speeches to Congress.Footnote ⁶ The Truman and Eisenhower doctrines were described in speeches to Congress, while the Bush Doctrine was codified in The National Security Strategy of the United States of America, published in September 2002.Footnote ⁷ The doctrines lacking this characteristic are those of Presidents Kennedy and Clinton. However, there are questions in both cases regarding whether any doctrine really existed. For example, Rabe (Reference Rabe2006) suggests that Kennedy was assassinated directly after declaring the vision that would have become a doctrine; hence, the policy was established during the tenure of the succeeding president and should, therefore, be named ‘the Johnson Doctrine.’ Furthermore, there are disagreements as to whether the Clinton Doctrine ever existed to begin with (Hyland, Reference Hyland1999: 203; Dumbrell, Reference Dumbrell2002: 43–56; Brands, Reference Brands2006: 4).Footnote ⁸ Beyond the realm of American foreign policy, the Soviet Union's Brezhnev Doctrine was presented by General Secretary Leonid Brezhnev at a political convention in Poland in November 1968, while the Japanese Fukuda Doctrine was proclaimed by Prime Minister Fukuda while visiting Manila, the Philippines, in 1977. Thus, all historically established foreign policy doctrines share this feature without exception.

For a set of beliefs to be considered an American foreign policy doctrine, they must contain the essential feature: a ‘clear declaration of policy at a particular place and time by a political leader.’ Meanwhile, questions remain regarding whether the Clinton Doctrine, which never involved a clear declaration of policy, was officially manifested. The Yoshida Doctrine does not exhibit this feature as there was no clear declaration of policy. Furthermore, at the conclusion of the Yoshida administration, there was no way to predict that Japan would simultaneously achieve minimal armament and economic growth. Thus, the Yoshida Doctrine is unique among foreign policy doctrines and, as a result, does not meet the standard of proof to be referred to as a ‘doctrine.’

Nagai and Nishihara, who advocated the Yoshida Doctrine, insisted that the doctrine has nothing to do with Yoshida's intentions or that such policies were never articulated to the public (Nishihara, Reference Nishihara, Bae-ho and Tadashi1978: 154; Nagai, Reference Nagai1985 [2016]: 258–259). Nonetheless, some researchers and several Japanese people continue to espouse the naïve assumption that Shigeru Yoshida determined Japan's post-war foreign policy. Part of the responsibility for this lies upon scholars from the 1970s and 1980s who carelessly used the unsuitable term and propagated its establishment and acceptance.

5. Yoshida Doctrine revisionism

The ‘Yoshida Doctrine’ does not possess the features that a foreign policy doctrine should. This is because the three pillars that form the doctrine were not announced in a clear declaration of policy by the Yoshida administration. However, even if Japan's foreign policy was not declared in the form of a doctrine, is the term ‘Yoshida Line’ suitable? Invoking the term ‘Yoshida Line’ as a conceptual term for Japan's post-war foreign policy is not suitable for two reasons. First, in the early-1950s US–Japan negotiations, Yoshida rebuffed calls for rearmament and laid the foundation for the ultimate strategy of minimal armament. These concepts are often posited as the rationale for the Yoshida Line, but research on Japan's foreign policy since the late 1990s has made it necessary to revise this assessment of the negotiations. Moreover, recent historical research has also revealed that Yoshida developed a proactive foreign policy toward mainland China. The second reason is that the Yoshida administration did not implement the policy of maintaining defense spending within 1% of the GDP to maintain minimal armament. On the contrary, the Yoshida administration spent more on the military compared to any other administration in Japanese history. These two points are examined in this section.

One issue that must be addressed before discussing the negotiations surrounding Japan's rearmament and actual defense spending is the fact that Yoshida did not plan for Japan to perpetually maintain minimal armament. As has been evidenced by prior research, Yoshida had no qualms about Japan becoming independent and, once it was financially feasible, establishing a national army (Ōtake, Reference Ōtake1988 [2005]; Hatano, Reference Hatano1989). Yoshida believed that military strength was a characteristic of a sovereign state. Yonosuke Nagai, who disseminated the term ‘Yoshida Doctrine’ in 1980, acknowledged that the doctrine was not the culmination of Yoshida's own beliefs (1985 [2016]: 259) but that Yoshida was unopposed to rearmament, and the doctrine was ‘unrelated to his own view.’ This fact was well known in 1980 when the phrase ‘Yoshida Doctrine’ emerged, as demonstrated by Prime Minister Yasuhiro Nakasone's remarks on the Lower House's Budget Committee on 2 February 1985. Referring to his efforts to strengthen the military by abolishing the defense spending limitation of 1% of the GDP, he stated, ‘If Yoshida were alive today, he would praise me for a job well done.’ Despite it not being Yoshida's intention, the three pillars (Figure 1) characterized Japan during and after Yoshida's tenure in the early-1950s, which is why Japan's limited armament is referred to as the ‘Yoshida Doctrine.’

The broadcast of this understanding of the Yoshida Doctrine can be attributed, as previously noted, to the alleged historical fact of the two times that Yoshida rebuffed American pressure to rearm Japan (Kōsaka, Reference Kōsaka1968; Nagai, Reference Nagai1985 [2016]; Pyle, Reference Pyle1987). The first occurrence was at the Dulles–Yoshida Talks in late-January 1951, and the second was at the Ikeda–Robertson Talks in October 1953 between Hayato Ikeda, Chairman of the Policy Research Council of the Liberal Party and former Minister of Finance, and Walter Robertson, the Assistant Secretary for Far Eastern Affairs. In his memoir, Yoshida wrote the following regarding the Dulles–Yoshida Talks:

However, this ‘historical fact’ required revision after the 1990s release of internal government documents, which led historians of foreign policy to critically reevaluate the Yoshida administration's foreign policy as well as Nagai's ‘Yoshida Doctrine’ and Kōsaka's assessment of Yoshida, in what could be called ‘the Yoshida Doctrine Revisionism.’ At the Dulles–Yoshida Talks in early-1951, Japan did not adhere to the rejection of rearmament. Rather, internal documents indicate that Japan agreed with the USA to begin rearmament, though to a limited degree, in the form of establishing a 50,000-person defense force (Uemura, Reference Uemura1995: 47–53; Miura, Reference Miura1996; Toyoshita, Reference Toyoshita1996: 60–64). Yoshida requested that Dulles and Douglas MacArther keep this establishment of a defense force confidential, and they agreed (Hatano, Reference Hatano1989: 195; Kusunoki, Reference Kusunoki2009: 223). This occurred confidentially, prior to the expansion and renaming of the National Police Reserve in October 1952, which led to a militaristic National Safety Force. After making this agreement with Dulles, Yoshida did not officially announce the rearmament. Very few people knew this secret; even the management of his own party was unaware (Uemura, Reference Uemura1995: 63).

Researchers are divided on whether this establishment of a security force should be understood as the result of American pressure to rearm (Muroyama, Reference Muroyama1992; Miura, Reference Miura1996; Toyoshita, Reference Toyoshita1996) or whether it should be interpreted as avoiding large-scale militarization and stopping at only a 50,000-person force (Tanaka, Reference Tanaka1997; Sakamoto, Reference Sakamoto2000; Kusunoki, Reference Kusunoki2009: 22). It is not the goal of this paper to evaluate the scale at which the Yoshida administration engaged in rearmament; however, it can be stated that the 1951 Dulles–Yoshida Talks, which were the major basis for the discourse on the Yoshida Doctrine, have been reconsidered.

Furthermore, since the 1990s, there have been major reassessments of the Ikeda–Robertson Talks of October 1953, another event used as rationale for the Yoshida Doctrine. At these talks, the USA pressured Japan, which was expected to enter a recession following the Korean War, to rearm as a condition for receiving aid through the US Mutual Security Act (MSA). Chairman of the Policy Research Council of the Liberal Party Hayato Ikeda met with Americans regarding MSA aid, and the two nations engaged in intense negotiations over the scale of Japan's rearmament. In response to the USA's demand for 10 divisions and 345,000 troops, Ikeda proposed 10 divisions and 180,000 troops, a proposition that was rejected. Ultimately, an exact number was not agreed upon, meaning that Japan had fended off American pressure for rearmament: ‘We are from that day [October 30, 1953] forward able to do away with the unfortunate impression that the US side with its large numbers was demanding a huge increase in Japanese defense forces’ (Miyazawa, Reference Miyazawa2007: 131). As Miyazawa, who would later become Prime Minister, later noted, they were understood as Japan's opportunity to cease American demands for rearmament. Documents from the 1990s have raised further questions. The Eisenhower administration, which had come to power during the same year, decided not to increase pressure on Japan to rearm; however, this was not revealed to Japan at the time of the talks because it was still undergoing an official review. Hence, the Ikeda–Robertson Talks were not the setting of the real negotiations regarding Japan's defense goals (Uemura, Reference Uemura1995: 145; Yoshitsugu, Reference Yoshitsugu and Toyoshita1999; Sakamoto, Reference Sakamoto2000).

The two ‘historical facts’ posited as a rationale for the Yoshida Doctrine have been re-evaluated because of the release of new historical materials, and at least one of its premises has been eliminated. In addition to defense policy, recent research has been reviewing the Yoshida administration's relations with other countries as well. One of them is the ‘counter infiltration’ plan against mainland China (PRC) proposed by Yoshida himself. Another case is Japan's involvement in the Korean War. As previously mentioned, one of the features of the Yoshida Doctrine is to avoid involvement in international political disputes. However, resent research revealed that Yoshida had a plan to send Japanese civilians to PRC as a ‘fifth column’ and to overthrow the government, proactively proposing the idea to the US government (Chen, Reference Chen2003; Inoue, Reference Inoue2009). Moreover, Yoshida proactively aimed to strengthen intelligence cooperation with both the USA and Taiwan. That plan did not come to fruition, but Yoshida was critical of US policy on China and thought that Japan could pull China away from the communist side. In addition, although limited, intelligence activities against China were also carried out in Japan (Inoue, Reference Inoue2009). Thus, newly revealed documents and historical studies that use those documents indicate the necessity of reconsidering one aspect of the Yoshida Doctrine – non-involvement in international disputes.

Moreover, what was revealed around 1980 is the fact that the Yoshida administration was directly involved in the Korean War and one Japanese died during the operation. Providing bases to the US army (the so-called UN forces) and selling supplies and services was a publicly known involvement in the Korean War. In addition to such indirect involvement, Japanese Coast Guard Special Sweepers were dispatched to the Korean Peninsula to engage in minesweeper for about 2 months from October 1950. The Japanese operation disposed of 27 mines, but one minesweeper sank after crashing with the mines, causing one death and 18 injuries (Suzuki, Reference Suzuki2005). In fact, this operation was not made public for the next 30 years (Wada, Reference Wada2014: 136–139). This is because this operation may violate Article 9 of the Japanese Constitution, and it is fully expected that it will become a political issue if it becomes public, so Yoshida instructed ‘to keep it secret at all’ (Suzuki, Reference Suzuki2005: 41). After the Korean War, Japan has avoided being involved in international conflicts until the Koizumi administration dispatched the Self-Defense Forces to the Indian Ocean (2001) and Iraq (2004) with the will of the USA. This is consistent with the view of previous research that the Koizumi administration ended the ‘Yoshida Doctrine.’ However, one parameter of the ‘Yoshida Doctrine,’ which does not involve in international conflicts, did not actually appear during the Yoshida administration. This fact should be a strong reason to question the post-war Japanese foreign policy named ‘Yoshida.’

6. Japanese foreign policies in the 1950s

As mentioned above, the ‘Yoshida Doctrine’ sprouted during the Yoshida administration, but it was the common understanding of previous research that it took root in the 1960s during the Ikeda and Sato administrations. In other words, Japan's diplomacy for about 6 years from the end of 1954, when Yoshida left the cabinet, is not called the Yoshida Doctrine or Yoshida Line. The reason is that Ichiro Hatoyama (Prime Minister 1955–1956), Tanzan Ishibashi (Prime Minister from December 1956 to February 1957), and Nobusuke Kishi (Prime Minister 1957–1960), who were in charge of the administration for those 6 years, were political rivals of Yoshida and had thought to have chosen different foreign policies from Yoshida, Ikeda, and Sato. Soeya (Reference Soeya1998), for example, argued that Hatoyama, Ishibashi, and Kishi chose the Autonomous Diplomacy vis-a-vis the USA, distinguishing them from Yoshida and Ikeda's collaboration with the US line. In addition, Hatoyama and Kishi have been categorized as ‘traditional nationalists’ and are argued to have been proactive toward rearmament (Ōtake Reference Ōtake1988 [2005]: 180; Iokibe, Reference Iokibe1989 [2005]). Ishibashi is also said to have been at the forefront of the rearmament debate, as the leader of the Hatoyama faction (Ōtake Reference Ōtake1988 [2005]: 191). In contrast, the understanding of previous studies is that Hayato Ikeda (Prime Minister from 1960 to 1964) and Eisaku Satō (Prime Minister from 1964 to 1972), who came to power in the 1960s, were senior members of the Yoshida faction and were, therefore, reluctant to rearm and thus encouraged the Yoshida Doctrine to commence.

This distinction, however, has also been significantly revised by recent research. Hoshiro (Reference Hoshiro2007), conducted a representative study of such revisions, and found that there was almost no difference between Yoshida, Ikeda, and Sato on the one hand, and Hatoyama, Ishibashi, and Kishi, on the other hand by comparing and examining five major foreign policies from the 1950s to the 1960s: (1) diplomatic recovery, (2) southeast Asian policy, (3) Japan–China and Japan–Taiwan relations, (4) rearmament policy, and (5) the Japan–US Security Treaty. The reason why the two were thought to be different was that, according to Hoshiro (Reference Hoshiro2007), researchers thought the mere policy proposals that Yoshida's rivals advocated when they were opposition parties or in the early days of the administration as their actual policies. However, in reality, those policy proposals of Hatoyama or Kishi rapidly faded away after they took office.

The transition of Japan's military spending from 1952 to 1960, after attaining independence provides an example for this argument. Figure 3 illustrates the ratio of defense spending to GDP, and Figure 4 shows the ratio of defense spending to national expenditures, while Figure 5 shows the actual amount of defense spending adjusted for inflation. The first 10 years following independence saw an escalation in Japan's military spending, which peaked in 1954 under the Yoshida administration (1952–1954). In the latter half of the Hatoyama administration, and in the Kishi administration, military spending as a percentage of GDP and the national budget was lower than it was at the end of the Yoshida administration. In Japanese post-war history, military spending exceeded 2% of the GDP only in the Yoshida administration and the first half of the Hatoyama administration (see Figure 2). Existing studies of Japanese diplomacy have only focused on Japan–US negotiations over rearmament in the early-1950s and have given little attention to comparisons of actual military spending between administrations. However, these data reveal that the administration with the relative highest military spending in the 1950s was the Yoshida administration and that the three subsequent administrations, which were Yoshida's political opponents, not Ikeda and Satō, rapidly reduced defense spending.

Figure 3. Japan's military spending as a percentage of GDP, 1952–1960. Source: SIPRI military expenditure database.

Figure 4. Japan's military spending as a percentage of the national budget, 1952–1960. Source: Zaisei chōsa-kai, ‘Kuni no Yosan [National Budget],’ various annual editions.

Figure 5. Japan's military spending (constant US million $, 2018). Source: SIPRI military expenditure database.

Incidentally, military spending in the 1950s can be divided into Japan's defense expenditures and defense contributions to the USA. The two were evenly divided under the Yoshida administration, and thereafter, with the gradual withdrawal of US forces, the share of the latter steadily decreased. Hence, under the Yoshida administration, rather than Japan's military expenditures being minimal, while protected by the USA, they were weighed heavily because of the presence of US forces.

Figures 3–5 reveal that in the 1950s, regardless of who became Japan's political leader, the amount of military spending never increased. In fact, the aggressive rearmament arguments of Hatoyama, Ishibashi, and others were all made before they assumed power, after which their arguments and the actual amount of military spending rapidly diminished, meaning the policy preferences of individual politicians did not influence the Japanese defense policies of the 1950s. Therefore, not only is it inappropriate to name a single politician, ‘Yoshida,’ as the motivator of Japan's low military spending, but it is also contrary to the fact that relative military spending, as a percentage of GDP and the national budget, peaked during his administration.

The above discussion and data suggest that, in the 1950s, regardless of who became Japan's political leader, its diplomatic lines remained largely unchanged. This brings the following two negative implications for the Yoshida Doctrine. First, proponents of the Yoshida Doctrine understand that the Yoshida Line was interrupted in the late 1950s and appeared in the 1960s again. However, this does not reflect the reality. Second, the policy preferences of individual politicians did not influence Japan's diplomacy in the 1950s. Hoshiro (Reference Hoshiro2007: 251) argues that during this era, structural factors such as the international environment (mostly US influence) and domestic budget constraints were so strong that they suppressed Japanese political leaders' original ideas about foreign policies of Japanese political leaders. Therefore, it is inappropriate to name a single politician, ‘Yoshida,’ because it gives the false impression as if Japan's foreign policies in the 1950s were the results of the free choice of a political leader.

7. Examination of the causal relationships that form the Yoshida Doctrine

7.1 Link with the US alliance, military presence, and military spending

One of the pillars of the ‘Yoshida Doctrine’ is the assertion that Japan's military expenditure was kept at a modest level because of choices made by the Yoshida administration, to conclude the Japan–US Security Treaty, and to provide bases. This view is compelling and sensible. Certainly, it is expensive to build a self-sustaining military, so it is fiscally responsible to use defense costs in other areas by forming an alliance with a major power and stationing their forces in-country. In fact, Shigeru Yoshida admitted that one of the reasons his administration urged the stationing of US troops in Japan in the early-1950s was that it could not spend such a large portion of the national budget on rearming Japan (Yoshida, Reference Yoshida1957 [1998]: 143). As a hermeneutic that reveals the intentions of the politicians of the era, the link between the alliance with the USA and provision of bases and the suppression of defense spending is valid.

Whether such an interpretation is a causal explanation of the real world, however, is another matter (Hollis and Smith, Reference Hollis and Smith1990). The consideration must be made as to whether this causal relationship has truly been substantiated, that is, whether the alliance with the USA and the provision of bases to US forces is the cause and the curtailment of domestic military spending is the result. For example, as shown in the previous section, the ratio of defense spending in the early-1950s was the highest in postwar history. As shown in Figure 6, the period with the highest number of US troops in Japan was from 1953 to 1954, at which point the US forces rapidly withdrew from Japan. There is no negative relationship between the number of US troops and military spending, but the data reveal that there is a positive relationship between the two, with the greater the number of US troops stationed, the greater the military spending as a percentage of the GDP. This is because defense contribution to the USA accounted for half of all military spending during the Yoshida administration. Incidentally, defense spending vanished in the 1960s but was revived in 1978 under the pretext of ‘Omoiyari Yosan (host nation support).’ The fact that Japan was funding the stationing of US troops and the fact that the USA was pressuring Japan to rearm in the early-1950s and the mid-1980s reveals that the USA was not entirely tolerant of its ally's restraint in military spending.

Figure 6. Number of US troops in Japan, 1950–2004. Source: Global US Troop Deployment, 1950–2003, by Tim Kane.Footnote ¹⁵

Table 2 shows the top 10 countries in which US troops stationed during each period from the 1950s to the 1990s (total number per decade). It can be confirmed that there are two patterns in the number of US troops stationed: countries such as West Germany and Japan, which had several permanently stationed US troops, and countries such as South Vietnam, which were neighbors of countries in conflict with the USA. The relationship between the number of US troops stationed and the amount of military spending is not always clear. During the Cold War, the defense budget of West Germany, which consistently had several US troops stationed in the country, was around 3% of the GDP, but after 1990, it was in the 1% range, declining as the number of US troops stationed in-country decreased (see Figure 2). France withdrew most of the US forces in the late-1960s but did not necessarily rapidly increase military spending after doing so (Figure 2).

Table 2. Number of US troops stationed around the world, 1950s–1990s

Source: Global US Troop Deployment, 1950–2003, by Tim Kane.

Japan's military spending in the early-1950s was high, though it had just signed the Security Treaty and hosted the largest number of stationed US troops. The original intentions of policymakers are not always realized; evidence indicates that Yoshida's expectations failed.Footnote ¹⁰ The ‘Yoshida Doctrine’ argument that the alliance with the USA and the provision of bases built by Shigeru Yoshida were the reasons Japan's military spending remained low do not reveal a causal relationship based on actual data. It is merely retrofit logic derived from the interpretation of the intentions of the Yoshida administration's policymakers and the historical fact that Japan's military spending was low, even following the 1960s, when Japan experienced profitable economic growth.

However, it is possible to argue that not only the number of stationed US troops but also the fact that Japan had an alliance with the USA is the cause of Japan's restrained defense spending. The best proof of that argument is to compare the case in which the possible cause exists and that in which it does not. However, both cannot be observed simultaneously. Only the case of the alliance with the USA and the provision of bases can be observed. Thus, rigorous verification is not possible, but comparative verification can be conducted using data from other countries. Therefore, this study examined the causal relationship between concluding an alliance with the USA and the provision of bases to ensure light equipment through a quantitative analysis using global datasets, including Japan. Specifically, the relationship between the two was analyzed using the level of military expenditure for each country as the dependent variable (result) and the following three as the main independent variables (causes): (1) whether the country has secured a defense alliance with the USA, (2) whether the alliance is asymmetrical, and (3) the number of US troops stationed in the country.

Findings of recent research have statistically identified the determinants of military spending (Dunne and Perlo-Freeman, Reference Dunne and Perlo-Freeman2003; Goldsmith, Reference Goldsmith2003; Dunne et al., Reference Dunne, Perlo-Freeman and Smith2008; Nordhaus et al., Reference Nordhaus, Oneal and Russett2012; Böhmelt and Bove, Reference Böhmelt and Bove2014). According to various studies, the factors that contribute to higher military spending include the size of a country's GDP, its economic growth rate, the undemocratic nature of the nation, and the fact that civil wars and conflicts between nations are taking place or are likely to take place. Financial inertia is at work, clarifying the previous year's military spending and its significant impact on the current year's spending. Both studies reveal that the number of allies does not lead to a reduction in a country's military spending. Rather, it has been reported that countries that have at least one alliance reveal higher military spending than those that do not (Albalate et al., Reference Albalate, Bel and Elias2012). Does constricting allies to the USA, the world's largest military power, affect military spending?

To answer this, the quantitative analysis in this study used the replication dataset of Böhmelt and Bove (Reference Böhmelt and Bove2014). These data are based on the bilateral (dyad) data of Nordhaus et al. (Reference Nordhaus, Oneal and Russett2012), with the unit of analysis changed to a single country (monad) and year (1952–2000). The logarithmically transformed amount of military expenditure constituted the dependent variable, and the following variables, which have been thought to affect the dependent variable, were the independent variables: (1) years since the last militarized interstate disputes (MID), (2) the POLITY democracy index score, (3) trade volume in GDP (export and import), (4) the number of countries bordering the country by land and sea, (5) the number of countries allied with the country in question, (6) the ratio of the country's GDP to the world's GDP, (7) the number of countries in existence that year, (8) the logarithmically transformed real GDP, (9) the logarithmically transformed military expenditure of the enemy nation, (10) the logarithmically transformed military expenditure, and (11) the dependent variable one period ago, that is, the logarithmically transformed military expenditure of the home country.Footnote ¹¹ However, Böhmelt and Bove (Reference Böhmelt and Bove2014) discovered that a model with only six of these 11 variables, (1), (2), (7), (8), (9), and (11), as independent variables does not change the results and is preferable because it is more conservative. Additionally, this study examined whether the following five independent variables affect the amount of military spending in each country: (1) a dummy variable that takes 1 if a country that has a defense alliance with the USA, (2) a dummy variable that takes 1 if an asymmetric alliance exists in which only the USA has a defense obligation, (3) the logarithmically transformed number of US troops stationed each year, (4) the interaction term between (1) and (3), and (5) the interaction term between (2) and (3). Alliance data use the Alliance Treaty Obligations and Provisions (ATOP) dataset for the Correlates of War (COW) and robustness checks (Leeds et al., Reference Leeds, Ritter, Mitchell and Long2002; Gibler, Reference Gibler2009).Footnote ¹²

A defense alliance is an agreement to jointly defend an ally in the event of an attack by a third country, and as of 2000, the USA concluded defense alliances with 55 countries in the COW dataset, with 56 in ATOP. Among the defense alliances, four countries in the COW dataset have experience with asymmetric alliances: Turkey, South Korea, Japan, and Pakistan. In addition to these four countries, ATOP includes Panama, Spain, Portugal, Iran, Iraq, Israel, and Uzbekistan. Hence, there is an asymmetric relationship in which the USA has a defense obligation to these countries, while these allies have no obligation to defend the USA.

The statistical model uses ordinary least squares (OLS) for panel-corrected standard errors (Beck and Katz, Reference Beck and Katz1995) because the independent variable includes the dependent variable one period earlier.Footnote ¹³ The dependent variable was logarithmically transformed defense spending.

The results of the statistical analyses are shown in Table 3. Thus, an alliance with the USA and the number of US troops stationed in a country do not affect the military expenditure of alliance partners. The implications of these results are thus clear. Even if a country forms a defense alliance with the USA, and the number of US troops stationed in the country is high, it does not necessarily lead to financial savings in the country's defense costs. Under the 1948 Vandenberg Resolution, the US defense alliance calls for ‘continuous and effective self-help and mutual aid’ from its partners so that the military expenditure of the other country is not suppressed by the alliance. Even in an asymmetrical alliance in which only the USA is unilaterally obliged to defend, military spending is not significantly different from other target defense nations as the USA demands the cost of stationing the US military and continues to press for its own defense. One of the pillars of the Yoshida Doctrine is the explanation that Japan's low defense spending is due to its alliance with the USA and the presence of US troops; therefore, it remains unproven according to statistical analysis of global datasets.

Table 3. Regression result (OLS with panel-corrected standard errors)

Notes. Panel-corrected standard errors in parentheses. All models include year fixed effects.

*P < 0.05, **P < 0.01, ***P < 0.001.

It is not the purpose of this study to clarify why Japan's military spending has been maintained within 1% of its GDP. According to previous studies that have discussed this, apart from the ‘Yoshida Doctrine,’ it is argued that the causes of the restraint in Japan's military expenditures were (1) the existence of the Constitution of Japan prohibiting the retention of military forces other than for self-defense, (2) rearmament not necessarily being supported by the public (culture of antimilitarism), (3) institutional sustainability since the Cabinet decision on the outline of the national defense program in 1976, (4) a strategy of buck-passing, where the need to balance against a threat is recognized, but it does as little of the required balancing as possible by relying on the efforts of others, and (5) the combination of three elements: pacifism, antitraditionalism, and the fear of entrapment (Berger, Reference Berger and Katzenstein1996; Hook, Reference Hook1996; Katzenstein, Reference Katzenstein1996; Chai Reference Chai1997; Lind Reference Lind2004; Izumikawa, Reference Izumikawa2010; Soeya, Reference Soeya2017). If their arguments are correct, and if those causes are removed, it can be expected that the possibility of Japan's military expansion will increase even if the Japan–US alliance survives.

8. Reconsidering the link between military spending restraint and economic growth

The argument that low-level armament enabled economic recovery and growth as another causal factor forming the ‘Yoshida Doctrine’ is also plausible. It is undeniable that Japan achieved substantial economic growth between the late-1950s and the early-1970s, which was unparalleled in the world at that time, with restrained military spending within 1% of its GDP. However, this simple causal relationship raises several questions. First, the link between restrained military spending and economic growth is merely a relationship that was limited up to the 1980s, when Japan's rapid economic growth was advancing. This is because Japan's economy has stagnated since the 1990s, even though military spending has remained within 1% of its GDP. This historical fact cannot be explained by the ‘Yoshida Doctrine.’ Second, no previous economic research has clarified the causal mechanism of how the government's suppression of military spending was a significant factor that led to Japan's economic growth. Saving as much as possible on defense costs from the national budget enabled financial investment in other areas, which the Yoshida administration or the Ikeda/Satō administration in the 1960s had already accomplished. However, Japan's fiscal policy was restrained during that period. Third, the ‘Yoshida Doctrine’ naturally assumes that there is an exchange between restrained defense spending and economic recovery and growth, but such a choice between the two is inappropriate. In fact, a so-called hegemonic state is one that has achieved both economic growth and arms expansion (Gilpin, Reference Gilpin1981; Modelski, Reference Modelski1987). Even in a country that is not a hegemonic superpower, for example, prewar Japan, the military industry shifted to the private sector, and economic growth and domestic technology development occurred simultaneously. Japan became a military and economic power, realizing ‘rich country, strong army’ (Samuels, Reference Samuels1994: chap. 3). There are also countries such as South Korea and Taiwan that experience high-military spending and high-economic growth rates as well as countries such as those in sub-Saharan Africa and Costa Rica that expend low or no military spending but experience low-economic growth rates. These facts suggest that curbing military spending is neither a sufficient condition for economic growth, where curbing military spending will always result in economic growth, nor a necessary requirement, where economic growth will not occur unless military spending is reduced.

9. Factors that contributed to Japan's economic growth

There are a large number of studies on political economy that discuss the factors that contributed to the success of the Japanese economy until the 1970s, but recent representative studies cite several factors as the reasons for achieving high-economic growth (Nakamura, Reference Nakamura and Nakamura1989; Yoshikawa, Reference Yoshikawa1997 [2012]; Takeda, Reference Takeda2011; Hara Reference Hara2012; Fukao, Reference Fukao2020).

First, it has been established that the following three were important initial conditions following the defeat and prior to the beginning of high growth. The first was the introduction of a Keisha Sesan Hōshiki (priority production system) from 1947 to 1948, in which limited resources were concentrated on coal and steel production. The second was the reactivation of the market mechanism through the 1949 policy of economic restraint (Dodge Line). The third was the demand for goods and services from the US military due to the Korean War, which began in 1950 (Chōsen tokuju, or Korean special procurements). It has become clear that Japan's highest growth continued from 1955 to 1972 due to the following factors: (1) a rising demand for private capital investment, (2) capital accumulation, (3) a long-term rise in labor productivity, represented by the rise in total factor productivity, (4) changes in the industrial structure (the effect of redistribution of labor between industries), (5) a mass creation of employment opportunities centered on the manufacturing industry, (6) an increase in self-employed commerce and industry, and (7) expansion of consumption represented by an increase in spending on durable consumer goods.

Of these, the one related to the ‘Yoshida Doctrine’ is (2), capital accumulation. When aggregate demand was extremely high, such as the case in Japan in the 1960s, maintaining high-speed capital accumulation required not only investment demand but also a large amount of savings supply. This was due to the sharp rise in private savings because of the income surge and the government's surplus finances due to the restraint in military spending (Nakamura, Reference Nakamura and Nakamura1989: 57–58; Horioka, Reference Horioka1990; Fukao, Reference Fukao2020: 200–201). However, the suppression of military spending did not directly cause economic growth through fiscal policy. This is because Japan's fiscal policy did not expand until 1970 (Ihori and Kawade, Reference Ihori and Kawade2001). Until then, Japan's fiscal policy was used as a market-restraining measure. For example, in 1957 and 1961, tightening policies were implemented, including the deferral of public works spending, fiscal investment and loan programs, and deferral of fiscal spending. The expansion of public works projects occurred in 1962, and in 1966, deficit-covering national bonds (construction bonds) were exceptionally issued, but this occurred only once, during the period of high-economic growth. From then until the 1970s, the Japanese government's standard policy was to maintain balanced public finance (Okazaki, Reference Okazaki1998).

Additionally, surplus public finance is effective for economic growth in providing capital resources only under the condition of extremely high aggregate demand, as in the 1960s. Following high-economic growth in the late-1970s, when there was a serious shortage of aggregate demand, due to a sharp decline in private investment, Japan had to deal with the problem of excess savings (Fukao, Reference Fukao2012). Naturally, efforts to restore the government's financial situation by reducing military spending did not provide a prescription for eliminating excess savings. Restraint in military spending had no impact on economic growth in Japan after the 1980s, when discourse regarding the Yoshida Doctrine was established.

10. Military expansion and economic growth

Studies on whether military spending affects economic growth have been established as defense economics, and the area has been expanding. Theoretically, there are contradictory positive and negative effects on the impact of a nation's defense industry on its economy (see Sandier and Hartley, Reference Sandier and Hartley1995; Smith, Reference Smith2009). The positive effects include job creation, infrastructure development, arms exports, effective use of unused capital, and technological spin-offs yielded by the defense sector to the private sector through defense R&D spending. The negative effects include the crowding-out effect (opportunity cost), where the national defense sector absorbs savings and taxation, investment in the private sector is prevented, and inflation rises due to the increase in the money supply. Economic effects can also be divided into short-term (Keynesian demand effects) and long-term (supply effects) effects. In this respect, the Yoshida Doctrine only considers the crowding-out effect among the various other effects. Nagai (Reference Nagai1985 [2016]: 78), for example, asserts that ‘if Japan had started moving in the direction of its own military industry and arms exports under the auspices of the MSA (Mutual Aid Agreement), the Japanese economic miracle of today would not have happened.’ However, this assertion is not the result of a systematic analysis of the relationship between military spending and economic growth but a conclusion drawn from the economic situation of the USA during the recession of the early-1980s when Nagai (Reference Nagai1985 [2016]) was written.

To date, the status of empirical studies with quantitative analysis has been mixed, and a common conclusion has not been reached among those who draw positive effects (Benoit, Reference Benoit1978; Atesoglu, Reference Atesoglu2002; Bose et al., Reference Bose, Haque and Osborn2007), negative effects (Lim, Reference Lim1983; Dunne and Tian, Reference Dunne and Tian2015, Reference Dunne and Tian2016; d'Agostino et al., Reference d'Agostino, Dunne and Pieroni2019), or no statistically significant relationship (Gerace, Reference Gerace2002; Heo, Reference Heo2010). The results of quantitative analyses also differ depending on the degree of economic development of each country (developed or developing), the international environment (presence or absence of external threats), and the degree of democracy in the country (Benoit, Reference Benoit1973; Aizenman and Glick, Reference Aizenman and Glick2006), and the results also differ depending on the model adopted by the researcher (Dunne et al., Reference Dunne, Smith and Willenbockel2005; Alptekin and Levine, Reference Alptekin and Levine2012). Furthermore, there are arguments that the relationship between the two is nonlinear, for example, an inverted U-shaped relationship (Stropu and Heckleman, Reference Stroup and Heckelman2001; Kalaitzidakis and Tzouvelekas, Reference Kalaitzidakis and Tzouvelekas2007). The analysis of the relationship between military spending and economic growth is still developing, and no unified view has emerged.

What is the relationship between military spending and economic growth if only Japan is considered? Mizuno et al. (Reference Mizuno, Ando and Igusa2016) provide the only study to theoretically derive the utility of Japan's defense. According to Mizuno et al. (Reference Mizuno, Ando and Igusa2016), the elasticity of scale in the utility function of the government is 1.1079, which is higher than that of the first order. Hence, Japan's defense budget, at within 1% of its GDP, is low. Increasing defense equipment and personnel could have significantly increased its utility. However, the Japanese government did not increase its military spending. In fact, the Japanese government opted to sacrifice the economy following the war to avoid aggressive security policies. For example, the Yoshida administration reduced the potential for technological spin-offs by abandoning the development of the arms industry (Samuels, Reference Samuels1994: chap. 3). Through the 1967 ‘Three Principles on Arms Exports’ or the ‘Unified Government View on Arms Exports (1976),’ the government abandoned the improvement of the trade balance through arms exports. Furthermore, even if there was a serious shortage of demand after the 1970s, it was not absorbed through military spending. Instead of contributing to the workforce during the 1991 Gulf War, it provided a massive $13 billion in funding. The source of the $13 billion included an increase in oil and corporate taxes, as the government was reluctant to contribute to the military workforce despite it slowed economic activity.

The above analysis and evidence indicate that the second causality of the ‘Yoshida Doctrine’ – that restrained military spending led to economic growth – is fallible.