Elite alignment plays a pivotal role in the consolidation of authoritarian rule (Grillo et al. Reference Grillo, Luo, Nalepa and Prato2024; Haggard and Kaufman Reference Haggard and Kaufman2021; Levitsky and Ziblatt Reference Levitsky and Ziblatt2018). Prevailing accounts emphasize that elites support authoritarian regimes when they are rewarded—through material benefits, policy influence, or institutional power—while sanctions and economic pressure are thought to push them in the opposite direction by raising the costs of loyalty. Yet such perspectives overlook moments when elites align not from opportunity, but from constraint. Politicians losing influence and firms weakened by sanctions may find resistance untenable, aligning with regimes out of dependence rather than conviction. This study bridges research on elite defection and economic coercion to show how material vulnerability, rather than ideological affinity or self-enrichment, binds elites to authoritarian governance.
Japan’s prewar political trajectory provides a valuable context for examining elite behavior under institutional strain. While not a full democracy, the country had implemented universal male suffrage by 1925, and the legislature retained meaningful veto power, often successfully challenging military or aristocratic authority. A substantial share of legislators—around 70%—also held business leadership roles, placing them at the intersection of political and economic power. By the late 1930s, as military influence expanded, formal institutions persisted but lost substance: accountability eroded, dissent was curtailed, and legislative checks on the executive steadily weakened. In this shifting environment, elite responses—whether resistant, accommodating, or passive—became increasingly consequential.
This study examines how material interests shaped elite responses to rising military influence in prewar Japan, focusing on legislators with corporate board ties to key economic sectors. It analyzes ten pivotal legislative episodes in which legislators’ attitudes toward the Imperial Japanese Army’s suppression of dissent were revealed—such as the motion to expel legislators critical of the military—highlighting how elites navigated the tension between defending institutional autonomy and accommodating authoritarian encroachment.
The analysis draws on two major economic shocks that disproportionately affected specific sectors. The first was the US sanctions (1940–41), which hit trade-dependent industries such as petrochemicals, textiles, and silk. The second was the expansion of military procurement, formalized in 1939 and broadened in 1942, which selectively benefited firms supplying the army and navy. Legislators tied to these sectors faced diverging economic incentives: some experienced acute disruption, while others stood to gain from state contracts. Tokyo Stock Exchange data show that these shocks were largely unanticipated, producing sudden and erratic price movements.
To assess how these shocks influenced elite behavior, this study employs difference-in-differences and event-study designs, using sectoral exposure as treatment and legislative behavior as the outcome. All models include legislator and event fixed effects.
The findings reveal a clear contrast: legislators from sanctioned sectors were significantly more likely to support military-backed policies after the shock, with robust results across specifications and no evidence of pre-trends. By contrast, legislators from procurement-linked sectors showed no comparable shift—and, if anything, became more distant from military actors over time—challenging the assumption that economic beneficiaries of war automatically back authoritarian rule.
A parallel case from the 1940 occupation of French Indochina further underscores this divergence. While rubber producers—beneficiaries of annexation—showed little political alignment, rice producers facing import competition exhibited clear authoritarian convergence. Nationally, legislators from sanctioned sectors were more likely to join the Imperial Rule Assistance Association (IRAA), a mass political organization closely aligned with the military, and to win seats in the 1942 election—even after losing financial support from traditional funders. At the individual level, the case of Kunitarō Koyama—a legislator from the sanctioned silk industry—shows how economic strain pushed him to realign politically and seek campaign funding from executive actors, abandoning earlier constitutional commitments. His trajectory exemplifies how diminished bargaining power, rather than ideological conviction, drove elite collaboration in the final years of competitive politics.
Alternative explanations—such as nationalist sentiment in response to external threats, fears of nationalization, or strategic delegation to a competent executive—are assessed using both quantitative and qualitative evidence. Yet none account for the timing or behavior of sanctioned legislators as convincingly as the bargaining power framework advanced here.
Although the sanctions were a nationwide crisis that might have prompted a rally-around-the-flag response, only legislators from sanction-exposed sectors realigned toward the military; procurement-linked and unaffected sectors remained stable. Moreover, the procurement shock—despite the regime’s appeals to national unity and heightened wartime fervor—produced no comparable shift among procurement-linked legislators. Together, these patterns highlight that economic distress, not patriotic conviction, underpinned elite collaboration with the regime.
These findings refine how we understand elite collaboration during authoritarian consolidation. They show that power asymmetries between elites and executives determine whether elites support or resist authoritarian initiatives. Regimes need not co-opt or repress all elites; securing a critical mass of vulnerable, easily co-opted elites can stabilize rule without broad repression. Weakened elites—especially those disadvantaged or losing under the status quo—often engage in passive collaboration, using authoritarian legislatures to negotiate rents and concessions (Opalo Reference Opalo2019; Reuter and Robertson Reference Reuter and Robertson2013). By contrast, stronger elites, whose positions remain secure, have fewer incentives to compromise and can use their bargaining power to demand checks that regimes cannot—or need not—grant.
This study advances the understanding of authoritarian transitions by showing how legislators and business leaders shape regime trajectories (Arriola, DeVaro, and Meng Reference Arriola, DeVaro and Meng2021; Grillo et al. Reference Grillo, Luo, Nalepa and Prato2024; Pospieszna and Vetulani-Cegiel Reference Pospieszna and Vetulani-Cegiel2021; Waldner and Lust Reference Waldner and Lust2018). Unlike voters, elites exercise shifting political and institutional leverage, making their alignment crucial during periods of military ascendancy and political uncertainty (Beliakova Reference Beliakova2021; Lorch Reference Lorch2021). The analysis also extends research on business–state relations and economic sanctions by linking sectoral exposure to elite behavior. Evidence from Japan shows that sanctions, meant to discipline regimes, can instead weaken internationally connected actors and shift the balance toward authoritarian consolidation (Afesorgbor and Mahadevan Reference Afesorgbor and Mahadevan2016; Szakonyi Reference Szakonyi2021).
LEGISLATIVE ELITES, HORIZONTAL ACCOUNTABILITY, AND AUTHORITARIAN SHIFT
Legislative elites are widely regarded as the principal guardians of horizontal accountability, expected to check executive power and preserve institutional balance (Waldner and Lust Reference Waldner and Lust2018). Yet they frequently fail to do so—a puzzle that sits at the core of democratic erosion (Miller Reference Miller2021). Unlike ordinary voters, legislators and other political elites possess both a direct stake in sustaining democracy and the strategic capacity to resist backsliding (Grillo et al. Reference Grillo, Luo, Nalepa and Prato2024). Their inaction, therefore, cannot be explained by popular apathy or weak institutions alone; it demands an account of why those best positioned to defend democracy instead allow it to erode.
Elite passivity is a central driver of democratic erosion, as political and economic actors often choose inaction when confronted with violations of democratic norms. While polarization and inequality are widely recognized as catalysts of authoritarian electoral shifts (Levitsky and Ziblatt Reference Levitsky and Ziblatt2018), elites frequently facilitate authoritarian consolidation through tacit support aimed at safeguarding their interests (Haggard and Kaufman Reference Haggard and Kaufman2021; Winters Reference Winters2011). Rather than initiating takeovers, they often allow backsliding to unfold unchallenged, thereby weakening the very institutions they are expected to uphold (Svolik Reference Svolik2013). This dynamic reflects Ermakoff’s (Reference Ermakoff2008) “paradox of abdication,” in which elected elites relinquish authority despite having a rational interest in preserving it. Business-aligned legislators, for instance, could plausibly safeguard their material and political interests by defending institutional constraints, yet their choice to dismantle them in alliance with opaque authoritarian forces remains a central puzzle. These dynamics extend beyond electoral democracies to hybrid regimes and civil–military relations (Lorch Reference Lorch2021).
Executive aggrandizement—where the executive gradually expands authority at the expense of other branches—often proceeds without elite resistance, accelerating the erosion of checks and balances and weakening judicial independence and legislative oversight (Bermeo Reference Bermeo2016; Levitsky and Way Reference Levitsky and Way2010). Historical cases such as Nazi Germany show how political and economic elites, whether through passive acquiescence or active support, enabled authoritarian consolidation to protect their status and assets (Ferguson and Voth Reference Ferguson and Voth2008; Wieviorka Reference Wieviorka2009).
To explain the breakdown of horizontal accountability, recent scholarship highlights how crises create strategic dilemmas for institutional “restrainers” (Gratton and Lee Reference Gratton and Lee2024; Howell, Shepsle, and Wolton Reference Howell, Shepsle and Wolton2023; Miller Reference Miller2021). Howell, Shepsle, and Wolton (Reference Howell, Shepsle and Wolton2023) show that emergencies can align the short-term goals of restraining institutions with those of the executive, prompting temporary transfers of authority that prove difficult to reverse. Gratton and Lee (Reference Gratton and Lee2024) similarly demonstrate how liberty–security trade-offs during crises compel restrainers to concede power, entrenching executive dominance. These dynamics echo historical cases such as wartime Britain, where Parliament delegated sweeping powers to the executive under the guise of crisis management.
Interwar Japan presents a compelling yet puzzling case. As a hybrid regime with a legislature capable of exercising veto power, it experienced the erosion of democratic checks and balances with the consent of many lawmakers. Yet this process was uneven—some resisted until the end, while others capitulated. This divergence, coupled with the military’s limited competence in managing the home front, challenges explanations that view delegation as a response to executive capability.
This study adopts a coalitional perspective on democratic breakdown (Waldner and Lust Reference Waldner and Lust2018), emphasizing power asymmetries and elite heterogeneity. Unlike voters, elites wield context-dependent political capital; for economic elites in particular, their shifting influence within democratic institutions shapes incentives to either defend or accommodate authoritarian expansion.
Politically weak or institutionally marginalized actors are especially prone to co-optation. Arriola, DeVaro, and Meng (Reference Arriola, DeVaro and Meng2021), in a cross-national study of African regimes, show that weak opposition leaders are more likely to be co-opted through strategic cabinet appointments. Timoneda (Reference Timoneda2021) argues that unproductive sectors, perceiving structural disadvantage, may align with anti-democratic coalitions, citing cases from the United States and Spain. Lührmann and Lindberg (Reference Lührmann and Lindberg2019), using global data, find that smaller party factions may support democratic erosion if they see it as a way to expand influence or protect partisan interests.
These studies establish the broader logic of selective incorporation within the repression–co-optation framework (Gandhi and Przeworski Reference Gandhi and Przeworski2006): weak elites are low cost to buy off and often receptive to integration. Because authoritarian regimes need not neutralize all opponents (Bueno de Mesquita et al. Reference de Mesquita, Bruce, Siverson and Morrow2003), targeting vulnerable elites to secure a critical mass of support becomes a cost-efficient strategy of consolidation.
Yet much of the existing literature treats elite weakness as a fixed attribute—focusing on who is co-opted rather than on how elites become co-optable. This approach risks conflating pre-existing marginalization with vulnerability that arises endogenously as economic and political conditions change, obscuring how established elites can become susceptible to co-optation when their material bases erode.
Building on this insight, the analysis focuses on how shifts in economic capital reshape political capital and, in turn, the cost of co-optation. Elites losing influence within the legislature during backsliding may risk forfeiting power and privilege. Yet authoritarian legislatures often endure as arenas for brokering rents and concessions rather than for democratic accountability or lawmaking (Reuter and Robertson Reference Reuter and Robertson2013). For elites already waning in democratic relevance—such as those lacking strong campaign finance—alignment with authoritarian coalitions can appear politically rational. For many legislators in hybrid regimes, particularly those tied to business interests, democratic institutions function more as instrumental tools than principled commitments, becoming expendable when the perceived returns to fighting for institutional survival are low.
In contrast, strong elites are often too costly to buy off and too dangerous to repress, so regimes tend to tolerate their independence while relying on weaker actors for political support. In Japan, for instance, some legislators continued to criticize the government until 1945 despite the government repression. Their persistence reflects how variation in elite power—who depends on the regime and who can resist it—shapes the course of authoritarian consolidation even within constrained institutions.
Selective co-optation—often described as a “divide and rule” strategy—fragments elite opposition and stabilizes regimes without full repression. This dynamic, coalitional view links material vulnerability to shifts in elite alignment, showing how targeted incorporation of weakened actors sustains authoritarian rule. Table 1 contrasts this framework with conventional theories of elite realignment and with delegation-based accounts of executive empowerment.
Table 1. Elite Responses to Political Capital Shocks: Theoretical Frameworks
ECONOMIC ELITES, SANCTION, PROCUREMENT, AND BACKSLIDING
The relationship between business and politics is well established (Bonardi, Hillman, and Keim Reference Bonardi, Hillman and Keim2005; Hertel-Fernandez Reference Hertel-Fernandez2016), but less is known about how legislators with business backgrounds behave politically, especially in their voting patterns (Krcmaric, Nelson, and Roberts Reference Krcmaric, Nelson and Roberts2020). Business influence is often strongest when private-sector actors hold key policymaking roles, such as central bank governors (Adolph Reference Adolph2013) or finance ministers (Jochimsen and Thomasius Reference Jochimsen and Thomasius2014). At the subnational level, Szakonyi (Reference Szakonyi2018) shows that Russian mayors with business ties used procurement to benefit their own firms, highlighting how personal economic interests shape policy behavior.
Few studies, however, examine how business-backed legislators position themselves toward democratic governance or backsliding. This study addresses that gap through prewar Japan, where most legislators also held business posts, offering a setting to test how economic interests shaped elite responses to authoritarian consolidation.
External economic shocks, however, can profoundly alter these relationships. Economic sanctions are a common tool for pressuring regimes that violate international norms, including democratic principles, yet their domestic effects are often complex and counterintuitive (Baldwin and Pape Reference Baldwin and Pape1998; Hufbauer, Schott, and Elliott Reference Hufbauer, Schott and Elliott2009). Intended to weaken authoritarian governments by constraining economic elites, sanctions can instead bolster regime stability through “rally around the flag” dynamics (Egger, Syropoulos, and Yotov Reference Egger, Syropoulos and Yotov2024) or create rent-seeking opportunities for politically connected actors. Sanctions on Venezuela’s oil industry, for instance, allowed regime-linked elites to profit from illicit smuggling networks, reinforcing authoritarian entrenchment (Bull and Rosales Reference Bull and Rosales2023).
While most research emphasizes macro-level outcomes, Afesorgbor and Mahadevan (Reference Afesorgbor and Mahadevan2016) show that export-oriented sectors bear disproportionate costs, whereas non-tradable sectors remain relatively insulated. This sectoral asymmetry produces divergent political incentives: some elites resist regime change to protect their interests, while others align with incumbents to secure protection. Despite the expansion of targeted sanctions designed to disrupt regime networks (Egger, Syropoulos, and Yotov Reference Egger, Syropoulos and Yotov2024), little is known about how business elites—as strategic actors—respond to such external pressures.
Sanctions on Japan before the Pacific War illustrate how external shocks interacted with parliamentary institutions, exposing fractures within the elite. Unlike modern theories that emphasize the deterrent or signaling functions of sanctions (Drezner Reference Drezner2003; Reference Drezner2022), the Japanese case represents one of the earliest comprehensive efforts to alter state behavior through economic coercion (Mulder Reference Mulder2022). It thus provides a valuable setting to examine how sanctions influenced elite alignment and institutional dynamics within a semi-democratic regime.
Conventional wisdom holds that sanctions generate economic pain that, in turn, produces public discontent and elite pressure for political change. Elites most exposed to sanctions are therefore expected to defend democratic institutions, since sanctions are meant to create incentives for affected groups to mobilize against authoritarian regimes (Marinov Reference Marinov2005).
This study proposes an alternative mechanism: sectors severely hit by sanctions or international isolation may become so weakened, politically and economically, that they cannot resist institutional erosion, making them more susceptible to co-optation.
A frequent reference point is apartheid-era South Africa, where sanctions are credited with spurring business opposition to authoritarian rule (Levy Reference Levy1999). Yet that resistance came mainly from large, resource-rich firms such as Anglo-American, whose financial security made reform advocacy low risk. It is less clear whether weaker firms—without comparable buffers—have played similar roles in democratic transitions.
Procurement represents the opposite dynamic. Unlike sanctions, government procurement programs—such as military contracts—allow regimes to co-opt economic elites by distributing material benefits. Albertus, Fenner, and Slater (Reference Albertus, Fenner and Slater2018) term this “coercive distribution,” whereby central authorities consolidate power or suppress dissent through selective resource allocation. Similarly, Rosenfeld (Reference Rosenfeld2020) finds that recipients of government benefits are more likely to support incumbents during episodes of backsliding. Procurement thus serves as a direct channel for aligning business interests with autocratic consolidation.
Yet the earlier discussion of legislative elites suggests a competing logic. While war profiteers might appear natural allies of authoritarian regimes (Fergusson and Voth Reference Ferguson and Voth2008), their dependence on the state can be reciprocal. Autocrats often avoid full nationalization—lacking capacity to manage firms efficiently—as seen in wartime Japan and Germany (Marshall Reference Marshall1967). Beneficiaries of procurement may instead use their leverage to resist centralization, seeking to preserve autonomy and protect long-term interests (cf. Hirschman Reference Hirschman1970). Even when materially advantaged, their existing influence under democratic institutions can diminish incentives to support authoritarian shifts.
In both sanction and procurement contexts, the adage “democracy costs money” is apt. Sanctions can deprive firms of the resources needed to sustain pro-democracy coalitions, weakening those most inclined to resist the regime. Procurement, by contrast, can bolster firms’ resources and bargaining power, enabling them to defend their autonomy against co-optation attempts by the state.
These dynamics yield two competing expectations, as outlined in Table 2. The conventional view holds that firms benefiting from the regime will support it, while those harmed will resist. This study instead argues that weakened sectors often lack the capacity to oppose authoritarianism, whereas stronger sectors—shielded by economic power—can afford to defend institutional autonomy.
Table 2. Sectoral Exposure and Expected Political Alignment
Ultimately, elite behavior reflects not only preferences but also political capital and capacity to act.
HISTORICAL AND INSTITUTIONAL CONTEXT
After its establishment in 1890, Japan’s House of Representatives (Diet) gradually expanded democratic participation, achieving universal male suffrage in 1925.Footnote 1 Although the emperor could appoint cabinets, military chiefs, and senior bureaucrats without legislative consent, all laws and budgets required Diet approval, giving it substantial influence over governance (Takenaka Reference Takenaka2014).
The military was constitutionally independent but depended on Diet approval for budgets and mobilization (Mimura Reference Mimura2011). Cabinets appointed by the emperor thus had to bargain with legislators for support. Even non-partisan cabinets such as those of Admirals Saitō and Okada (1932–36) secured parliamentary majorities only through concessions to party factions.
Despite political assassinations and instability in the early 1930s, the Diet continued to convene regularly, debate policy, and scrutinize government actions until 1941. Political repression remained limited and largely targeted communists and anarchists (Shillony Reference Shillony1991).
Electoral institutions further reinforced the Diet’s pluralism. Elections operated under a single non-transferable vote (SNTV) system, with three to five representatives per district. Because multiple co-partisans competed within the same district, party cohesion was weak, and candidates relied heavily on personal networks and business sponsors rather than party platforms. Many legislators held corporate board positions, allowing business interests to mobilize electoral support and influence policy (Masumi Reference Masumi1985; Pekkanen Reference Pekkanen2006).
In this context of fragmented but still competitive politics, the Diet frequently challenged military influence. The Katō and Hamaguchi cabinets, backed by business-friendly parties, reduced army and navy budgets substantially in the 1920s and early 1930s. In 1937, General Hayashi dissolved the Diet after failing to pass a pro-army budget but won only 40 of 466 seats, forcing his resignation. Legislative opposition also contributed to the downfall of the Hirota (1937), Konoe (1939), and Abe (1940) cabinets.
By the early 1940s, however, this balance collapsed. The creation of the IRAA in 1940 dissolved all political parties to support the war effort, though limited opposition persisted through new parliamentary groups. In the 1942 election, non-IRAA-endorsed candidates still captured about 35% of the vote and 18% of seats, and elections remained relatively free from coercion (Shillony Reference Shillony1991, 12). Yet by the war’s end, the Diet had become largely ceremonial, passing nearly all government bills with minimal debate as the army authorities pre-cleared parliamentary speeches.
During this period, the army lacked cohesive leadership and suffered from internal divisions and command dysfunction—as seen in the occupation of French Indochina and the failed attempt to nationalize key industries (discussed in the following sections). Mid-level officers often acted without central approval, yet the army’s political mandate nonetheless expanded, strengthening the executive branch.
Rather than abolishing competitive elections, the army and its allies targeted anti-military politicians through selective repression and electoral disadvantage (Furukawa Reference Furukawa2001).
Within a decade, Japan’s political landscape transformed. In 1936, legislators still competed for power and exercised veto authority; by 1942, political parties had disappeared, and the Diet largely ratified policies that curtailed its own power. This paradox—how legislators, many tied to business interests and invested in their own autonomy, nevertheless yielded to military pressure—lies at the core of this study.
DATA COLLECTION AND PRELIMINARY RESEARCH DESIGN
This research examines how Japanese legislators responded to the army’s efforts to undermine democracy, using a difference-in-differences approach. The dependent variables are legislators’ actions and voting behavior during key moments of military influence. The independent variables include legislators’ affiliations with industries affected by exogenous economic shocks—either negative (due to sanctions) or positive (through military mobilization). To account for individual legislator traits and time-specific shocks, the models use both legislator and event fixed effects, with full details provided in the research design section below(for replication data, see Fukumoto Reference Fukumoto2026).
Voting behavior and parliamentary affiliations were collected from parliamentary minutes and endorsement lists for resolutions, proposals, and questions (Shūgiin Jimukyoku 1928–45). While the minutes are publicly available online, the endorsement lists are housed at Waseda University Library.
Biographical data were compiled from Jinji Kōshinroku (Jinji Kōshinjo 1937–43) and the Shūgiin Yōran (Shūgiin Jimukyoku 1937, 1942), both available in the Japanese Diet Library. These datasets were digitized using OCR software (Google Cloud Vision and ABBYY FineReader) and verified by multiple research assistants for accuracy.
Biographical information includes place and date of birth, education, titles, prior occupations, political career, and military service. The dataset also records whether a legislator belonged to an aristocratic family, political dynasty, or was adopted into one. Political career variables capture prior service as local or prefectural councilor, mayor, cabinet member, committee chair, party executive, or member of the IRAA. Legislators’ economic interests are proxied by board memberships in private firms or business associations, as listed in Jinji Kōshinroku, which distinguishes between current board memberships and former occupations. These affiliations are coded as binary indicators representing legislators’ connections to specific industries.
Economic shock timing and affected sectors were identified using Kabukai Nijūnen (Tōyō Keizai Shinpōsha 1937–43) and Kabushiki Nenkan (Nomura Tokushichi Shōten 1937–44), both archived at Waseda University Library. From Kabukai Nijūnen, I extracted sectoral stock index summaries reporting end-of-month values by industry, which were digitized and standardized into consistent time series for cross-sectoral comparison. Kabushiki Nenkan provided complementary quantitative and qualitative information on firm- and sector-level disruptions, including government interventions, and offered a more detailed industrial classification used to refine sectoral categories in the analysis.
Government procurement data were obtained from the army’s Shitei Kōjō Sagyōjō Meibo (Ministry of War 1940, 1942) and the navy’s Gyōshu Betsu Kaigun Kanri Kōjō Meibo (Ministry of Navy 1942), both housed at the Ministry of Defense Archive. They contain comprehensive lists of procured factories and ownership information. Sample images of archival sources and digitized materials are provided in Appendix I of the Supplementary Material.
DEPENDENT VARIABLES: ANTI-DEMOCRATIC LEGISLATIVE ACTIONS
The army’s repeated violations of Japan’s democratic institutions—especially after martial law was imposed in 1936—undermined parliamentary oversight and suppressed dissent. This study defines the dependent variable as legislators’ cross-party resistance or acquiescence to army-backed proposals.
After the failed 1936 coup, the military’s institutional power expanded. Active-duty officers were again required for army and navy minister posts, and the new Military Affairs Bureau (Gunmukyoku) oversaw defense policy beyond military affairs.Footnote 2
The coup was framed as a response to political corruption, and the military-led cabinet adopted the slogan “renewing civilian administration.” The new Cabinet Research Bureau and Military Affairs Bureau led anti-political-party campaigns blaming corruption for the unrest.Footnote 3
The army’s anti-democratic stance became explicit in late 1936. On October 23, the Military Affairs Division proposed administrative restructuring, declaring that “given the rejection of party politics based on majority rule, the system will inevitably assume the form of a one-party state.”Footnote 4 A week later, the Tokyo Asahi Newspaper reported plans to restrict suffrage to household heads and veterans and to introduce a “Political Party Law” limiting party activity in the Diet—measures discussed in a five-minister council on parliamentary reform.Footnote 5
Army factions and Interior Ministry allies, despite internal rifts, shared hostility toward parties and civil liberties. On April 5, 1937, Interior officials resolved to suppress criticism of military expansion,Footnote 6 foreshadowing the later fusion of military and bureaucracy, embodied by Tōjō’s dual War–Interior ministerial posts and General Abe’s leadership of the IRAA.
Legislators initially resisted, blocking army-backed bills on military secrecy and electricity nationalization and decisively defeating pro-army candidates in the 1937 election. Yet within 5 years, the army accomplished what repression alone could not—subduing the very majority that had once defied it. This study investigates how that paradox unfolded.
A key challenge in this research is the limited documentation of legislators’ stances on anti-democratic cabinet proposals. Controversial bills rarely reached open votes, as the Diet typically blocked them in committee. For example, the Hayashi cabinet—widely seen as anti-democratic—introduced 38 bills, only 11 of which passed, while the remaining 27 lapsed without formal rejection.
In contrast, legislator-initiated bills, motions, resolutions, and questions were far more likely to reach open votes. These proposals circulated in advance and included endorsements listing supporting parliamentarians. Such endorsements serve as the primary evidence of who supported or opposed the suppression of legislative authority and democratic norms. Additional insights come from records of pro-army legislators who split from moderate factions in 1937, 1939, and 1940, with affiliations documented for each session.
This study examines six instances where legislator-tabled activities revealed individual political positions and four cases where parliamentary group splits and reorganizations reflected attitudes toward democratic institutions. Pro-army, anti-democratic positions are coded as 1, and anti-army, pro-democratic positions as 0, with detailed coding explained in Table Y2 in Online Appendix Y of the Supplementary Material. Brief descriptions of each event appear in Table Y2 in the Supplementary Material, and fuller accounts are provided in Appendix K of the Supplementary Material. Analyses limited to either legislator-tabled activities or parliamentary group splits are presented in Appendix D of the Supplementary Material.
All resolutions, motions, questions, and pieces of legislation analyzed in Table Y2 in Appendix Y of the Supplementary Material share a common thread: they capture cross-party resistance—or collaboration—in suppressing dissent and weakening democracy.Footnote 7 Resolution 70-11/-12, Legislation 73-19/-20, and Legislation 74-28/-31 were rare moments of cross-party unity against military and police repression. On these occasions, multiple parliamentary groups submitted identical proposals on the same day, seeking unanimous endorsement—though none fully achieved it. By contrast, symbolic resolutions, such as those praising soldiers or commemorating imperial anniversaries, consistently passed unanimously. Most other initiatives in the early period reflected partisan divides shaped by proposers’ affiliations, making these episodes exceptional instances of cross-party cooperation before the 1940 party dissolution.
Unlike resolutions or legislation, motions were seldom used in the 1930s, when party discipline reduced their necessity. Their use surged after 1940 as the army penetrated parliamentary procedures and weakened party influence, turning motions into battlegrounds between the army and the Diet. Motion 75-1 supported the army’s push to expel an anti-army member; Motion 76-1/-2 and Questions 76-5/14 challenged unconstitutional subsidies to pro-army legislators; and Motion 77-1 sought to shield Prime Minister Tōjō from critical questioning.
Some legislative initiatives of the period may appear self-serving, but this does not contradict the paper’s central claim. Oversight, transparency, and tolerance of dissent were also institutional strategies through which business-aligned legislators advanced their own material interests. The Prosecutors Office Bill (Legislation 73-19/-20), often caricatured as retaliation against prosecutors enforcing election laws, in fact reflected growing concern over prosecutorial overreach—especially after politically charged cases widely regarded as fabricated. The 1937 Teijin Incident, later discredited as a false-flag scandal aimed at discrediting elected politicians,Footnote 8 exemplified these fears. The bill thus represented not mere self-protection but a defense against unchecked judicial power.
Although party discipline in Japan was weak, cross-party collaboration on proposals and questions remained rare until 1939. Yet, the party splits that did occur consistently centered on attitudes toward the army and parliamentary democracy. In April 1937, Prime Minister General Hayashi’s attacks on party resistance to the military led to the Diet’s dissolution and splintering within all major parties. By 1939, further fragmentation produced openly pro-army factions—such as the Nakajima and Yanai groups—breaking away from moderates.Footnote 9
In 1940, the army-backed Konoe cabinet created the IRAA to establish a one-party state, absorbing cooperative members from all factions and dissolving the existing parties. Dissenters formed new opposition groups—Dōkōkai (Liberal) and Kōa Dōmei (Hawkish but anti-IRAA). The 1942 election marked another turning point: the army, through the IRAA, formally endorsed candidates and financed their campaigns. Yet this alignment was not entirely coercive—legislators strategically sought or rejected IRAA backing, revealing the calculated nature of their collaboration (Furukawa Reference Furukawa2001).
To address concerns about event selection bias, placebo tests using other resolutions, proposals, questions, motions, and open votes from the 76th–79th parliamentary sessions (between the party dissolution and the 1942 election) are reported in Figure F1 in Appendix F of the Supplementary Material. The six cross-party legislations analyzed in the main text represent the full set of identifiable instances in which legislators collectively opposed the army; most other initiatives during this period were partisan in nature.
A related concern is that the selected events may appear idiosyncratic and difficult to compare. Because few bills were ever reintroduced, truly identical cases are unavailable. However, these ten episodes form a coherent set: each captures cross-party positioning toward army encroachment on the legislature, observed at different stages of democratic erosion. The sequence thus traces how resistance and accommodation evolved—from legislative defense of civil oversight to procedural struggles over parliamentary autonomy—as the army consolidated control. While the cases differ in form, together they provide the most systematic and comparable evidence available for examining Japan’s gradual democratic breakdown.
The coding process is detailed in Table Y1 (Japanese) and Y2 (English) in Appendix Y of the Supplementary Material. Legislators who died, were expelled, suspended, or recalled were excluded, limiting the sample to those who attended parliamentary sessions. No legislators were assassinated or executed during the study period.
The erosion of Japan’s semi-democracy is often attributed to institutional weakness (Berger Reference Berger1977), yet institutions themselves were endogenous and contested. The events analyzed here show that both the judiciary and legislature were active arenas of struggle, not passive constraints. Legislative efforts—such as Legislation 73-19/-20—helped strengthen judicial independence, and the courts repeatedly ruled the IRAA unconstitutional, even ordering a reelection in Kagoshima’s 2nd district in 1944. Conversely, the army’s veto power over cabinet formation was restored in 1936 amid post-coup turmoil. These dynamics reveal that institutional change reflected ongoing power struggles between the executive and legislature, rather than predetermined weakness, underscoring the value of examining discrete political events.
SUMMARY STATISTICS: PROFILE OF LEGISLATORS
Figures J1, J2, and J3 in Appendix J of the Supplementary Material summarize data on 1,086 legislators serving between 1936 and 1942, the period in which the Diet gradually ceded its authority. The average legislator was 50.8 years old, and 14% did not survive the war.
As Figure J1 in Appendix J of the Supplementary Material shows, 69% of legislators held executive or board positions in private firms, reflecting the close integration of business and politics.Footnote 10 Roughly one-third also belonged to industrial or employers’ associations, though sectoral details are often unclear. Social interest groups such as labor unions and agricultural associations involved roughly 21–24% of legislators, with considerable overlap. About 69% attended university, and professions such as law, teaching, journalism, and the bureaucracy were well represented, each accounting for roughly 14–18% of legislators. Samurai or aristocratic lineage, or second-generation legislators, accounted for only 6% and 4%, respectively.
Figure J2 in Appendix J of the Supplementary Material presents the distribution of business affiliations by sector, following the 1937 Kabushiki Nenkan classification. Two additional categories—Sanctioned and Procured sectors—identify industries most affected by international sanctions and wartime procurement. While newspapers and banks featured prominently, representation was broadly dispersed across sectors, with no single industry dominant. Overall, 17.4% of legislators were tied to sanctioned sectors, while roughly 10% had interests in procured firms. Only the electricity sector underwent partial nationalization.
Table 3 shows that traditional indicators of ideology or personal background—such as party affiliation, military service, or bureaucratic experience—did not consistently predict support for the army by 1942. Authoritarian alignment thus cut across partisan and occupational lines.
Table 3. Army Endorsement: Background of Incumbents
Figures L1 and L2 in Appendix L of the Supplementary Material visualize legislative networks based on shared backgrounds. In Figure L1 of the Supplementary Material, dark squares denote independents opposing the army, while light circles those endorsed by it in 1942; links represent shared education, regional origin, or career ties. The network shows no strong clustering by endorsement, suggesting that personal and regional ties had limited influence on democratic alignment. This finding challenges the conventional view of Japan’s prewar Diet as a personalist institution dominated by factional or friendship-based politics.
Figure L2 in the Supplementary Material maps connections based on shared business or interest-group ties. A weak clustering of independents in the lower-left quadrant suggests that economic affiliations—rather than personal networks—played a more important role in shaping legislators’ political alignment.
Table 4 shows no pre-1936 partisan bias between legislators in sanctioned and procured sectors, confirming adequate pre-treatment balance.
Table 4. Legislators Present in 1936: Initial Party Balance Test
Figures 1 and 2 present t-test results for business categories and pro-army attitudes in the 1937 and 1942 elections, covering legislators who served throughout the 1937–42 term. Business executives were mildly anti-army in 1937 but shifted toward pro-army positions by 1942. Those linked to procured sectors remained largely stable and anti-army, whereas legislators tied to sanctioned sectors exhibited the sharpest shift—from anti- to strongly pro-army—over the same period.
Figure 1. Business-Related Variables: T-Test Results for 1937 General Hayashi Coalition
Figure 2. Business-Related Variables: T-Test Results for 1942 Army Endorsement
EMPIRICAL ANALYSIS 1: ECONOMIC SANCTION
This section examines how economic sanctions and trade disruptions affected legislators from tradable sectors, focusing on whether they supported or opposed the army’s consolidation of power before and after the sanctions.
Following Japan’s military expansion in China and Southeast Asia, Western powers—particularly the United States—imposed successive trade restrictions, summarized in Table M1 in Appendix M of the Supplementary Material. Early measures by the League of Nations, Britain, and the Netherlands had limited impact, but two major US embargoes produced sharp market shocks. In response, the government banned short-selling and directed banks to purchase stock indices to stabilize the Tokyo exchange.
Figure O1 in Appendix O of the Supplementary Material shows the Tokyo Stock Market Index from 1937 to 1942, as reported in Kabukai Nijūnen (Tōyō Keizai Shinpōsha 1924–44). Solid lines indicate US sanctions; dashed lines mark key war events: the Marco Polo Bridge Incident (1937), Germany’s invasion of Poland (1939), the fall of Paris (1940), and the attack on Pearl Harbor (1941). Despite interventions, the market fell sharply after the US embargoes, while other wartime events produced smaller reactions and no comparable policy responses (Kabushiki Nenkan, Nomura Shōten 1937–44).
The timing of the first major US sanction is crucial for causal inference. The September 1940 trade embargo followed Japan’s occupation of northern Indochina—an operation that was not centrally planned. In early September, during ongoing negotiations to secure transit rights through French Indochina, Army Operations Chief Kyōji Tominaga prematurely ordered troop mobilization without cabinet approval. Although Tokyo had reached an agreement with Vichy France on August 4, Lieutenant General Akito Nakamura unilaterally advanced troops on September 23, sparking a brief two-day clash. The incident provoked outrage within the Navy and Army High Command, leading to the dismissal of Tominaga and other senior officers.Footnote 11
Because the operation was unauthorized and widely condemned as a failure of military discipline, the ensuing US embargo—imposed three days later on September 26, 1940—was viewed as an unforeseen consequence rather than a predictable outcome of policy choice. From the legislators’ perspective, this makes the sanctions plausibly exogenous. Following the stock market collapse, the Japanese government banned short-selling and raised deposit requirements on October 4 to stabilize financial markets.
Japan’s occupation of southern Indochina on July 21, 1941, by contrast, was a deliberate diplomatic move negotiated with Vichy France and carried out without conflict. Yet the US response five days later, on July 26—a full asset freeze and oil embargo—again stunned policymakers.Footnote 12 Contemporary accounts describe disbelief that Washington would act so decisively, given its earlier leniency toward Germany after France’s fall. Industry groups likewise assumed that trade would continue via neutral intermediaries, as it had with China, according to the 1942 Textile Yearbook (Nihon Sen’i Kenkyūkai Reference Kenkyūkai1942). The Japanese government again instructed major banks to intervene and stabilize the stock market on July 28.
Sectoral Analysis of Economic Sanction
Monthly reports in Tōyō Keizai highlight sharp sectoral disparities following the US embargo. Even export-oriented industries such as textiles suffered from bans on oil and iron, compounded by the contraction of yen–dollar trade. While few sectors were spared, tradable and resource-dependent industries bore the brunt—most notably steel, petroleum, and petrochemicals. Although Japan sourced some iron domestically and from its colonies, dependence on imports placed severe strain on steel firms, while petroleum shortages worsened even in occupied regions before 1942.Footnote 13
Tables N1 and N2 in Appendix N of the Supplementary Material summarize Japan’s trade composition before and after the sanctions. In 1938, imports were dominated by raw materials—most notably raw cotton, which accounted for nearly one-quarter of total imports—while exports centered on finished textile goods. Textile products made from cotton, silk, and nylon occupied six of the top export categories. By 1942, total exports had fallen by roughly half: cotton textiles declined to one-tenth of their pre-sanction levels, and raw silk exports nearly ceased. Occupied territories failed to replace lost Western markets, underscoring the severe disruption to Japan’s export-oriented industries (Tōyo Keizai Shinpōsha Reference Shinpōsha1940–42).
Figure O2 in Appendix O of the Supplementary Material plots sectoral stock indices from Kabukai Nijūnen (Shōwa 12–18 editions, 1937–42). Export- and import-oriented sectors consistently underperformed, with sharp declines after the sanctions. Petrochemical stocks suffered the steepest losses, while textile shares also fell markedly, reflecting the combined effects of trade restrictions and supply shortages.
Although most legislators’ firms were unlisted, their economic interests likely followed these market patterns. Small- and medium-sized firms were closely tied to major sectoral actors and thus faced similar shocks.Footnote 14
Sectoral indices were compiled by Tōyō Keizai at an aggregate level. Coverage varied across industries—textiles were fragmented across multiple listings, while steel and trading firms were few and grouped with related sectors—so observed differences reflect both market performance and reporting structure.
Figure O3 in Appendix O of the Supplementary Material shows that insurance companies’ share prices, despite the asset freeze, were largely unaffected—likely due to their domestic client base. In contrast, stock exchange and brokerage firms suffered steep declines in market valuation, reflecting dependence on foreign capital, while banks experienced only mild losses.
The worst-performing category was the Miscellaneous index, which included trading firms heavily engaged in international commerce, although detailed firm-level composition is not fully documented.
Drawing on these patterns, I classify the following as sanction-affected sectors: textiles (cotton, silk, synthetic), petroleum and petrochemicals, steel, stockbrokers, and international trade—five of the six weakest indices in this period. Insurance, though underperforming, shows no sanction-specific effect and is therefore excluded. Each sector is analyzed separately in robustness checks.
Empirical Specifications on Economic Sanction
This section uses a difference-in-differences and event-study design to analyze the political behavior of legislators affected by sanction shocks, compared to those unaffected. The individual legislator is the unit of analysis.
To estimate the effects, all models incorporate legislator fixed effects and event (date/legislation) fixed effects. Standard errors are clustered at the individual legislator level to account for potential correlations within individual records. The primary specifications can be expressed as
$$ {y}_{k,t}={\gamma}_k+{\delta}_t+\beta \hskip0.1em \left({Sanctioned}_k\times {Post}_t\right)+{\epsilon}_{k,t}, $$
where
$ {y}_{k,t} $
is a binary outcome indicating whether legislator
$ k $
adopted a pro-army position in event
$ t $
. As noted in the previous section, pro-army attitudes are observed through recorded votes, endorsements, factional alignment, and sponsorship or opposition to motions directly related to the army’s attempts to weaken parliamentary authority.
$ {\gamma}_k $
are legislator fixed effects, and
$ {\delta}_t $
are event fixed effects.
$ {Sanctioned}_k $
equals 1 if legislator
$ k $
held a corporate board position in a sector defined as sanctioned in the previous section, and 0 otherwise. Legislators without such ties form the comparison group.
$ {Post}_t $
takes 1 if the event occurred after the onset of sanctions.
Earlier sanctions imposed by the League of Nations, Britain, and the Netherlands had limited market impact. September 26, 1940 marks the peak of the initial sanction shock, with the sharpest stock market declines observed in sanctioned sectors, and is therefore used as the cutoff. Treatment is applied uniformly from this point onward, rather than staggered, because all sanctioned sectors experienced severe disruption simultaneously. The sectors affected again after the 1941 sanctions were already under strain by late 1940, as shown in Figure O2 in Appendix O of the Supplementary Material.
A “sanction shock” is thus defined as the sudden exposure of a legislator’s affiliated sector to embargoes that produced measurable market disruption. In some specifications, I exclude the two events between the initial embargo (September 1940) and the full and final US sanctions of July 1941 to create a cleaner pre/post contrast.
Event-study models apply the same specification, using 1937 as the baseline to capture long-term dynamics and demonstrate the absence of systematic pre-trends.
Potential confounding shocks—such as the Battle of Khalkhin Gol or the proposed nationalization of the electricity sector—are examined in later sections. These posed localized or sector-specific risks but did not cause the same widespread disruption as the US sanctions. Appendix D of the Supplementary Material further presents robustness checks restricting either the outcome variables (e.g., legislation only) or the legislator sample (e.g., board members only) to confirm that results are not driven by coincident shocks.
Results on Economic Sanction DiD
Due to space constraints, the event-study figures documenting pre-trends and dynamic treatment effects are reported in Appendix A of the Supplementary Material; these figures are central to the identification strategy and closely track the difference-in-differences estimates reported below.
Table 5 reports the difference-in-differences results. Legislators tied to sectors affected by economic sanctions became significantly more pro-army and anti-democratic after the sanctions took effect. Given the binary outcome, the coefficients (0.140–0.168) are large in magnitude. Models 2 and 4 exclude mid-sanction events (September 1940–July 1941), while Models 3 and 4 restrict the sample to legislators who participated in all ten events. Results remain robust across specifications.
Table 5. Summary of Difference-in-Differences Results with Two-Way Fixed Effects: Sanction (1940.9)
Note: *p<0.05; **p<0.01; ***p<0.001.
Figure A1 in Appendix A of the Supplementary Material presents the event-study estimates. The figure traces pro-army attitudes among legislators linked to sanctioned sectors and confirms the dynamics underlying the DiD estimates: a sharp post-September 1940 shift toward pro-army positions, with no discernible pre-trends. This pattern strengthens the causal interpretation, showing that the sanctions were the first major economic shock to affect sectors asymmetrically. The two sanction dates—September 1940 and July 1941—are marked with dotted lines.
Figure A2 in Appendix A of the Supplementary Material disaggregates results by sector. Except for international trade—where stock data are incomplete and legislator behavior appears erratic due to sectoral volatility—patterns closely mirror the main findings. Sectors directly hit by sanctions and suffering sharp market losses, notably petrochemicals and steel, show the strongest pro-army shifts. The leading export sector, cotton textiles, shows a mild pre-trend but turns sharply pro-army after 1940, especially relative to the broader textile group. Small-scale oil sellers, excluded from prioritized supply, also display a pronounced shift compared to protected oil refiners. Overall, the sanctions mark a clear turning point, as legislators from exposed sectors realigned away from anti-army, pro-legislative positions.
Figures A3 and A4 in Appendix A of the Supplementary Material show placebo tests for unaffected sectors. As expected, insurance—whose stock prices were stable—shows flat results, as do most non-tradable or domestically oriented industries. Non-tradable sectors such as services, retail, transport, and construction exhibit no discernible movement following the sanction shock. Similarly, sectors not dependent on international trade—including paper, fertilizer, ice, and forestry—remain stable. Only legislators linked to sanction-exposed sectors display a clear realignment, underscoring that the observed shifts are driven by economic exposure rather than general wartime sentiment.
Figure A5 in Appendix A of the Supplementary Material examines banks, whose shares declined modestly after the sanctions. Their pattern is weaker but parallels that of sanctioned sectors. To probe further, legislators are divided into two groups: those connected to mutual banks (small, local institutions) and those linked to commercial or industrial banks with international exposure. Only the latter show a significant pro-army shift, underscoring that vulnerability to foreign asset freezes shaped political realignment.
Overall, the evidence shows that exposure to sanction-induced economic distress was a key catalyst in shifting legislators from affected sectors toward pro-army, anti-democratic positions.
EMPIRICAL ANALYSIS 2: WARTIME PROCUREMENT
This section examines the political behavior of legislators whose businesses were affected by wartime procurement, assessing whether they supported or opposed the army’s consolidation of power.
Following the outbreak of the Second Sino-Japanese War in July 1937, Japan faced severe resource shortages that disrupted both the economy and military operations. Although the army sought control over key weapons factories, the preamble of the 1942 Navy Designated Factories and Plants List highlights its limited administrative capacity: even among Class A (army-controlled) factories, one officer was responsible for supervising three sites (Ministry of Navy 1942).
During this period, there was no nationalization or expropriation of private firms, nor did the army or navy place representatives on corporate boards. Instead, firms were compensated for revenue losses caused by restricted civilian production, with compensation calculated from the previous three years of financial records (Ministry of Navy 1942).
To analyze how procurement affected legislators’ political alignment, I digitized the 1942 Navy Designated Factories and Plants List and the 1940/1942 Army Procurement Contractors Lists archived at the Ministry of Defense. Together, they identify 2,428 firms contracted by the military. Legislators were coded as “procured” if they held board positions in firms supplying goods or materials to the army or navy. The 1940 list covers only army suppliers, with 868 additional firms appearing by 1942—indicating that most procurement relationships were established well after the 1937 war began.
Before the procurement system was reorganized in 1939, following the outbreak of World War II in Europe, contracts were mostly negotiated ad hoc without a centralized framework (Miwa Reference Miwa2015). Despite Japan’s declared neutrality, procurement expanded substantially after 1939.Footnote 15
Certain sectors, such as pharmaceuticals, automobiles, and heavy industry, were disproportionately represented in the procurement lists compared to firms listed on the Tokyo Stock Exchange (Table P1 in Appendix P of the Supplementary Material). As these sectors do not entirely overlap with the procured firms in the legislator dataset, they are analyzed separately.
Because data on early-stage procurement are limited, the precise timing of shocks cannot be identified with certainty. As shown in Figure O4 (Appendix O of the Supplementary Material), sectors linked to procurement exhibited little or even negative market reaction to the 1937 mobilization, reflecting the army’s limited industrial reach at the time. The informal 1938 list included only 358 firms—about one-seventh of the 1942 total—indicating that military involvement in industry was still minimal. Accordingly, the 1937 shock is excluded from the difference-in-differences analysis.
By contrast, stock prices in related sectors rose sharply after Germany’s invasion of Poland in August 1939 and Japan’s attack on Pearl Harbor in December 1941, marking the rapid expansion and formalization of procurement programs, clearly observed in Figure O4 in the Supplementary Material. Both developments were unanticipated, providing plausibly exogenous shocks for analysis. Consequently, greater emphasis is placed on the event-study specification, which captures these dynamics and allows assessment of pre- and post-trends.
Empirical Specifications: Wartime Procurement
This section employs a difference-in-differences framework, analogous to the previous analysis, using legislators with interests in mobilized sectors as the treatment group. Treatment is defined by board memberships in firms designated by the army or navy for procurement:
$$ {y}_{k,t}={\gamma}_k+{\delta}_t+\beta \hskip0.1em {Procured}_k\times {Post}_t+{\epsilon}_{k,t}, $$
where
$ {Procured}_k $
is a dummy variable that takes one if the legislator
$ k $
experienced mobilization shock during the period of this study, and
$ {Post}_t $
takes 1 if the vote or other relevant event t is after the occurrence of the shock. As in the previous analysis,
$ {y}_{k,t} $
is an outcome dummy variable, pro-army attitude shown by legislator
$ k $
at an occasion
$ t $
.
$ {\gamma}_k $
are legislator-level fixed effects,
$ {\delta}_t $
are event fixed effects.
Given the limited precision of early procurement data, the analysis focuses on the unanticipated expansions following Germany’s invasion of Poland (1939) and Japan’s attack on Pearl Harbor (1941). The event-study specification is emphasized, as it best captures these shocks’ timing and pre- versus post-trends.
Empirical Results: Wartime Procurement
Tables 6 and Figure B1 in Appendix B of the Supplementary Material (event-study graph) show no systematic pro- or anti-army shift among legislators linked to firms in the 1942 procurement programs, despite the substantial business opportunities these contracts created with the army and navy.
Table 6. Summary of DiD Results with Two-Way Fixed Effects: Procurement
Note: *p<0.05; **p<0.01; ***p<0.001.
In Table 6, Models 1 and 3 use the German invasion of Poland (August 31, 1939) as the shock, while Models 2 and 4 use Pearl Harbor (December 7, 1941). Columns 3 and 4 restrict the sample to legislators present at all ten events analyzed in this article. Across specifications, no discernible effects are found for procurement-linked legislators, even though their sectors’ stock prices surged after both events. Figure B1 in the Supplementary Material display both shock dates with dotted lines; treatment is not staggered.Footnote 16
Figures B3 and B4 in Appendix B of the Supplementary Material show a modest downward trend among legislators tied to industrial procurement sectors, such as automobile and industrial painting, suggesting a mild anti-army drift. Because the parallel-trends assumption is violated, these results are only suggestive and should be interpreted with caution. Figure B2 in Appendix B of the Supplementary Material focuses on industrial firms only, comparing legislators connected to firms on the procurement list with those linked to non-procurement firms. Despite noisy pre-trends, the overall pattern mirrors the sectoral analysis: procurement-linked legislators drifted slightly away from the army after the shock, while others moved closer.
Overall, the null findings across models indicate that legislators tied to procurement firms were not significantly more likely to align with the army during the studied period.
Empirical Results: Sanction vs Wartime Procurement
Figure 3 presents the event-study graph for a combined (“horse-race”) analysis comparing legislators linked to procurement firms, those in sanctioned sectors, and those connected to both. These effects are estimated simultaneously. The results show a pronounced pro-army shift among legislators tied to sanctioned sectors, especially after the first wave of sanctions.
Figure 3. Event-Study Graph for Procured, Sanctioned, or Procured and Sanctioned Firms
These findings challenge the conventional assumption that beneficiaries of military expansion were the regime’s strongest supporters. Instead, legislators representing sectors damaged by international isolation were more likely to align with the army, whereas those in procurement-linked industries—despite material gains from the regime—showed little change in political stance.
ROBUSTNESS CHECKS
The main difference-in-differences and event-study models include two-way fixed effects and clustered standard errors, limiting the scope for additional controls. Table D1 in Appendix D of the Supplementary Material reports subsample analyses that isolate either parliamentary factions or legislative actions within the pro-army score, avoiding direct comparisons between factional splits and roll calls. Across all specifications, results remain positive and statistically significant. Additional robustness checks in Table D2 in the Supplementary Material—such as excluding socialist legislators, Tokyo representatives, former cabinet ministers, or legislators without business board memberships—produce similar results.
Figure C1 in the Supplementary Material tests whether background or ideological traits—such as military or bureaucratic service, veteran or colonial ties, foreign study, or higher education—account for pro-army alignment. Across all panels, no systematic associations emerge: by 1942, elites from every background behaved similarly. The absence of ideological or institutional effects underscores a critical point—personal identity, career experience, and education did not predict authoritarian alignment. Instead, alignment followed patterns of material vulnerability, confirming that economic exposure, rather than conviction or institutional loyalty, drove elite realignment.
Figure E1 in Appendix E of the Supplementary Material shows no clear geographic clustering of legislators connected to sanctioned or procured sectors, or of pro-army alignment itself. Legislators from procured sectors were somewhat concentrated in industrial prefectures but remained widely dispersed, including in rural regions. A Kolmogorov–Smirnov test (Figure F2 in Appendix F of the Supplementary Material) finds no significant age differences across groups.
Appendix G of the Supplementary Material examines whether residual party loyalties shaped legislative behavior after party dissolution. Network analysis of co-voting patterns in the 76th–79th parliaments (1940–42) shows no clustering among former Minseitō or Seiyūkai legislators—the two largest prewar parties—indicating that informal party discipline had effectively collapsed.
Appendix H of the Supplementary Material analyzes the 1937 and 1942 elections using lasso regression to assess the relative influence of sanctions and procurement exposure. Related factors strongly predict pro-army alignment in 1942 but not in 1937, underscoring that economic shocks, rather than inherited partisan orientation, drove the realignment.
Discussion on Mechanism 1: Parallel Analysis with Annexation of French Indochina
The paradoxical findings warrant closer scrutiny. Economically weakened sectors may have aligned with authoritarian forces through several channels—the most intuitive being co-optation: distressed groups were cheaper to buy off, while thriving sectors were costlier to capture. The evidence points in this direction.
Japan’s annexation of French Indochina, which precipitated the sanctions, was undertaken primarily for strategic rather than economic reasons—aimed at cutting China’s supply routes rather than securing new markets or resources. Consequently, its direct economic impact was limited, minimizing concerns about confounding shocks while still generating modest, sector-specific effects. The rubber industry, for instance, benefited and avoided the shortages that afflicted other strategic sectors.
By contrast, the rice sector suffered. The Abe cabinet’s 1939 rice price controls proved difficult to enforce amid resistance from agricultural associations.Footnote 17 After 1940, cheap Indochinese imports depressed unofficial domestic prices, weakening the bargaining position of rice farmers (Ishikawa Reference Ishikawa1941, 32). The Teikoku Nōkai Nenpō (Imperial Agricultural Association Annual Report) expressed alarm that colonial expansion had deepened mainland price declines (Sawada Reference Sawada1941, 29–31).
No single interest group in the Diet represented rice farmers alone. The Imperial Agricultural Association (AA) functioned as an umbrella organization encompassing multiple commodity unions—covering livestock, tea, tobacco, forestry, sericulture, and others—each representing their respective producers. Because rice, by far the country’s largest agricultural product, lacked a dedicated union, legislators affiliated only with the general AA or with the Land Cultivation Union (Kōchi Kyōkai) are likely to have been primarily engaged in rice farming.
To test whether these disruptions shaped political alignment, I re-estimate the difference-in-differences model, focusing on sectors affected by the Indochina annexation. Figure Q1 in Appendix Q of the Supplementary Material shows no political shift among rubber manufacturers—mirroring the null results for procurement beneficiaries—but a clear pro-military drift among legislators affiliated with the general Imperial Agricultural Association (AA) or the Land Cultivation Union (Kōchi Kyōkai). In contrast, legislators belonging to specialized AA sub-unions (e.g., livestock, tea, or sericulture) show no comparable change.
Although this test is coarse—lacking precise data on Indochinese import volumes, black-market rice prices, or legislators’ direct exposure to the rice economy—it closely parallels the main findings. Economic vulnerability, rather than ideology, emerges as the central driver of elite realignment.
Discussion on Mechanism 2: Imperial Rule Assistance Association Funding
Although officially designated a non-political organization, the IRAA actively supported pro-military candidates in elections after its foundation in 1940. It received generous public funding, and staff salaries reportedly tripled those of ordinary public servants. While such funds were not legally intended for campaigning, strong evidence indicates they were used for that purpose (Furukawa Reference Furukawa2001). In addition to its annual budget of 8,300,000 yen, the IRAA obtained an unaccounted 930,000 yen from the Second Reserve Fund in 1941, with no public disclosure of how it was spent.
During the 76th Imperial Diet session (February 1941), Etsujirō Uehara delivered one of the most forceful parliamentary critiques of wartime authoritarianism. He submitted a formal motion to stop the government’s allocation of 930,000 yen from the Second Reserve Fund to the IRAA. He argued that reserve funds—intended for unforeseen emergencies such as natural disasters or external shocks—were unjustifiably diverted to a politically motivated organization lacking any legal foundation. Uehara condemned the Association’s structure as unconstitutional, warning that by embedding military officers, prosecutors, and bureaucrats within its ranks, it subordinated the legislature to the executive and was evolving into a one-party authoritarian regime resembling Nazi Germany, the Soviet Union, or even a bakufu-like dictatorship. He ridiculed the government’s claim that the IRAA was a mere “public association,” noting the absurdity of official comparisons to hygiene cooperatives.
Uehara further cautioned that freedom of speech was disappearing even within the Diet itself, where only IRAA affiliates could speak without fear. “Without criticism and reflection,” he warned, “neither constitutional government nor national well-being can be sustained.”Footnote 18 Accepting this precedent, he argued, would destroy parliamentary sovereignty.
Although his motion was defeated (54–192), the unusually high number of abstentions (196) reflected significant unease within the Diet over the nature of the organization. Appendix S of the Supplementary Material presents excerpts of his speech.
Legislators linked to sanctioned sectors were significantly more likely to join the IRAA (Figure R1 in Appendix R of the Supplementary Material), while procurement-linked sectors showed no such tendency. Candidates from sanctioned sectors also had a higher likelihood of winning in the 1942 general election (Figure R2 in Appendix R of the Supplementary Material), despite economic hardship facing their key funders. These patterns strongly suggest that political support was secured through material inducements. The case studies that follow reinforce this interpretation.
Discussion on Mechanism 3: Case Study of Rep. Kunitarō Koyama
This section draws on documentary evidence showing that Kunitarō Koyama—a legislator previously aligned with party democracy—received IRAA funding to secure his electoral survival, lending further support to the argument that economically or politically vulnerable candidates were effectively “bought off.”
Autobiographies by Japanese legislators are rare, and even those who held ministerial posts left few records of their wartime activities. This silence likely reflects postwar politics: many purged legislators returned to office after 1952, giving them incentives to avoid documenting collaboration. As a result, systematic case selection is impossible; this section instead illustrates potential mechanisms.
Kunitarō Koyama (1889–1981) is one of the few business-backed legislators whose legacy is well documented through two biographies by his local supporters (Inosaka Reference Inosaka1979; Publication Committee for Omoide no Kunitarō Sensei 1982). His political legacy, carried on by his son-in-law, grandson, and great-grandson—who also became legislators—likely motivated efforts to commemorate his contributions.
Born into a landowning family whose father served as mayor, Koyama graduated from a local commercial school, completed military service, and inherited a small silk-weaving firm in Nagano. Through his management, it grew into a prosperous exporter to the United States—an achievement that propelled his rise to local and national politics as mayor, prefectural councilor, and, by 1928, member of the House of Representatives.
He repeatedly advanced bills and petitions promoting silk price stabilization and production expansion, directly reflecting his stake in the industry. In 1934, for example, he submitted Proposal 65-97 Regarding the Regulation of Imported Used Silk Stockings to shield domestic producers from foreign competition.
Early in his career, Koyama publicly emphasized legislative independence and electoral integrity. On February 10, 1929, he delivered a speech criticizing government interference in elections and led an impeachment debate against the Minister of Justice. While this episode partly reflected party politics, it also revealed his early concern for parliamentary accountability. Nonetheless, outside these interventions, his primary focus remained on promoting silk industry interests.
According to his two biographies, Koyama—then at the height of his business success—was a vocal opponent of the army’s interference in the private economy and consistently prioritized the export market. This is corroborated by his February 25, 1937 speech, in which he criticized the government’s Raw Silk Price Stabilization Facility Act as harmful to export competitiveness and denounced emerging totalitarian economic policies.
As a member of the Yonai Cabinet in 1940, Koyama advocated conciliation with China and worked toward a ceasefire with Chiang Kai-shek, even traveling to Hainan Island to negotiate despite obstruction from the military. He was also among the few cabinet members to oppose the army’s attempt to expel Representative Takao Saitō (Inosaka Reference Inosaka1979, 175–81).
Economic sanctions marked a turning point in Koyama’s priorities. Although he had earlier opposed centralized measures, he joined the Silk Textile Ration Committee in late 1941. Around the same time, despite his 1929 criticisms of government interference in elections, Koyama joined the IRAA and remained silent as the government suppressed dissent. In the dataset, he is coded as pro-army in four post-sanction events in this research, compared to his anti-army stance in six prior events. His reelection in 1942 with army endorsement further underscores this transition.
In his memoir of Koyama, Inosaka records that Koyama’s secretary, Yūtarō Kutsukake, received 2,000 yen in campaign funds from the IRAA’s secret budget. According to Kutsukake, Koyama refused to accept the money directly, suspecting it originated from military secret funds. Nonetheless, Kutsukake handed the 2,000 yen to Koyama’s chief campaign staff (Inosaka Reference Inosaka1979, 212). Other candidates reportedly received up to 4,000 yen—enough at the time to purchase a house in Tokyo. Inosaka also recounts a conversation with prefectural councilor Amane Miyashita, one of Koyama’s supporters who had recommended him to the IRAA. Miyashita argued that “budget matters more than pride” for the sake of the silk industry and implied that without IRAA financial support, Koyama would likely lose his seat (Inosaka Reference Inosaka1979, 209).
After the sanctions, Koyama made several notable speeches beyond his usual silk policy agenda—on the National Rehabilitation Fund Act (1942), the Enterprise Restructuring Finance Provision Act (1943), and in Budget Committee discussions (1944). Each reflected his efforts to secure government support for struggling textile firms, emphasizing refinancing and restructuring to mitigate sanction impacts. In September 1944, he proposed Proposal 85-26 to defend silk resources during the war and successfully negotiated the exclusion of his firm and others from the failed centralization of the silk distribution network.
A contrasting case study of Torakichi Nakano—a legislator tied to procurement industries who followed the opposite trajectory—is presented in Appendix U of the Supplementary Material.
Koyama’s trajectory exemplifies the core argument of this study: economic strength profoundly shaped legislators’ stances and their alignment with the army. When his business prospered, Koyama resisted military interference and supported parliamentary autonomy; under economic distress, he aligned with the army, prioritizing political survival and business continuity over democratic oversight.
Discussion on Alternative Mechanisms 1: Nationalism and Rally-Around-the-Flag Effect
An alternative explanation could be the rally-around-the-flag effect or a broader rise in nationalism and patriotism. Sanctions are often associated with this phenomenon (Egger, Syropoulos, and Yotov Reference Egger, Syropoulos and Yotov2024), which may well have occurred in Japan during the 1940s. The rally-around-the-flag effect—typically defined as a surge in public support for national leaders during periods of external threat—generally presupposes efforts by political elites to emphasize national unity over partisan or sectional interests. Had the government primarily sought to rally support through appeals to national crisis, responses would likely have been uniform across sectors rather than varying with business characteristics. The behavior of procurement-linked sectors—arguably the most receptive to nationalist appeals—suggests that any rise in nationalism was uneven and likely limited rather than universal.
While a broad rally-around-the-flag effect appears unlikely, a more selective rise of nationalism among legislators from sanction-hit sectors remains possible, perhaps reflecting anger or a sense of national trauma rather than opportunism. In Shayo’s (Reference Shayo2009) framework, such behavior could reflect a shift in self-identification toward the national collective when economic status or group standing declines. Assessing these identity-based reactions is difficult, as personal memoirs from this period are scarce and often contested. Instead, I examine parliamentary behavior during the Koiso cabinet era (1944–45), when legislative activity briefly revived and provides a systematic window into ideological realignment. During the 85th Imperial Diet session in September 1944, legislators introduced 28 proposals. Most addressed practical wartime concerns such as food production, logistics, and insurance, but two—one in particular—displayed overtly nationalistic themes.
On September 9, 1944, a motion titled “Proposal Concerning the Establishment of Imperial National Morality” was introduced in the Imperial Diet. It called for a sweeping ideological mobilization of the state and society to support the war effort. Declaring Japan a divine nation (shinkoku), the motion urged national unity to achieve a new world order grounded in imperial morality and “proper guidance” from the state to concentrate the people’s energy toward total victory. It concluded by demanding moral sanctions—such as seppuku, exile, or house arrest—to enforce political accountability and strengthen the war footing. Appendix T of the Supplementary Material presents excerpts.
Figure V1 in Appendix V of the Supplementary Material shows that legislators linked to sanctioned sectors were not especially likely to support the proposal on imperial national morality, suggesting their alignment was not driven by ideological right-wing conviction.
By contrast, Proposal No. 9—introduced in the same session and calling for the armament of all civilians—was less ideologically charged and faced more opposition from procurement-linked sectors, likely due to concerns over labor shortages, while legislators from sanctioned sectors again showed no strong preferences (Figure V2 in Appendix V of the Supplementary Material).
It is difficult to make a definitive statement, since milder measures—such as resolutions to thank soldiers or honor the war dead, which might better reflect nationalist sentiment—were almost universally supported across sectors and over time. Nevertheless, legislators from sanctioned sectors do not appear to have been particularly ideological compared with others in 1944, two years after the 1942 election. The evidence thus suggests that their alignment was primarily transactional rather than ideological.
Earlier accounts, such as Marshall (Reference Marshall1967), emphasized an ideological fusion of capitalism and nationalism among business elites. By contrast, this study finds that political alignment was driven less by ideology than by sectoral economic vulnerability.
Discussion on Alternative Mechanisms 2: Business Consolidation and National Control
As we discuss wartime Japanese activities, one might assume extensive government-controlled consolidation during the war. However, most sectors remained under private ownership, with wartime government-backed cartels still largely run by private firm owners.Footnote 19 The military did not appoint representatives to corporate boards either.
In December 1940, with strong backing from Minister of War Hideki Tōjō, the Planning Agency announced the Outline for Establishing a New Economic System, which aimed to implement a tightly controlled wartime economy.Footnote 20 However, strong opposition from the business community led to the removal of key provisions, including the proposed “separation of capital and management” and strict limits on dividends. The subsequent arrests of 17 Planning Agency and IRAA personnel on charges of harboring state-socialist tendencies further weakened the push for economic centralization.Footnote 21 Figure W1A in Appendix W of the Supplementary Material shows a negative but minor reaction in the stock market following the announcement. No parliamentary events analyzed in this study occurred during this episode of policy turbulence, which ultimately underscored the army’s limited capacity to nationalize or control the civilian economy.
Legislators tied to sectors under relatively strict government control did not exhibit corresponding shifts toward anti-democratic or pro-army behavior. Two sectors subjected to stronger national control were electricity and liquor, yet neither underwent full nationalization. Hydroelectric plants and liquor distilleries were excluded from nationalization, and shareholders of private firms retained stakes in the resulting national monopolies, partly thanks to the pushback from related legislators.Footnote 22 Figure W1B in Appendix W of the Supplementary Material shows that electricity-sector stocks fell briefly after the partial nationalization but did not experience a persistent downward trend. Legislators connected to these sectors showed no consistent political tendencies (Figure W2 in Appendix W of the Supplementary Material), indicating that government control alone did not produce uniform shifts in alignment.
From late 1941 to early 1943, 24 control associations were formed under imperial ordinances as state-backed cartels. Administrative overlap and weak authority meant they achieved little, leaving private firms largely autonomous.Footnote 23 These associations postdated the sanction shocks and are unlikely to affect the main results.
Discussion on Alternative Mechanisms 3: Conflict in Manchuria
The Battle of Khalkhin Gol (summer–fall 1939) occurred within the study period, introducing a potential complication. As a border conflict, it did not directly damage major industries but created uncertainty for Manchurian businesses. Figure X1A in Appendix X of the Supplementary Material shows that Manchuria-linked firms experienced moderate stock market declines, reflecting this heightened risk. Their capital and operations, however, were largely unaffected by the fighting. Other theaters in mainland China remained relatively static during the sanction period and did not generate comparable shocks. Legislators tied to Manchurian firms display only a weak pro-army tendency—observable but far less pronounced than among sanction-affected sectors, as shown in Figures X1B–D in Appendix X of the Supplementary Material. This pattern supports the broader argument: firms able to withstand or sidestep disruptions faced weaker incentives for political realignment.
Discussion on Alternative Mechanisms 4: Salience and Delegation
Another alternative mechanism is that some legislators—particularly those from vulnerable sectors—may have deferred to the military not because of coercion or material inducement, but out of a belief that wartime conditions required greater flexibility and centralized authority. From this perspective, the war effort was paramount, and granting the military wider discretion was seen as essential for national success (Gratton and Lee Reference Gratton and Lee2024).
Yet pro-army alignment in the Diet did not necessarily translate into passive delegation. Many legislators who shifted positions also assumed new executive roles in ministries, especially in areas of domestic governance. This blurs the line between strategic delegation and institutional alignment. Military command and operational planning were already beyond parliamentary purview; instead, the Diet’s main responsibilities lay in budgeting, economic management, and resource mobilization—domains in which pro-military legislators exercised active influence rather than merely relinquishing authority. At least four legislators were dispatched to govern occupied territories in Southeast Asia, further illustrating the military’s dependence on civilian politicians.
The competence-based interpretation is also difficult to sustain. As the Saitō case illustrates, the army’s prolonged campaign in China drew increasing parliamentary criticism, weakening its perceived legitimacy. The efforts by the military to nationalize key industries were scaled back in the face of bureaucratic and business opposition, highlighting the military’s limited policy know-how and political capital in economic affairs.Footnote 24 In matters such as supervising military procurement factories, the army lacked sufficient personnel to manage the home front and frequently delegated responsibilities to civilian legislators.
Procurement-linked legislators—those most embedded in the war economy—did not exhibit a comparable pattern of deference, suggesting that issue salience alone cannot account for the observed alignment. The outcomes examined in this study primarily capture institutional deference, such as the curtailment of interpellation, rather than support for military strategy itself. While salience-based explanations cannot be ruled out, the evidence more strongly supports the view that alignment reflected material vulnerability rather than ideology or beliefs about competence.
CONCLUSION AND IMPLICATIONS
While this research does not dispute the influence of weak democratic institutions and culture in Japan’s authoritarian transition, the fact that most legislators defended the semi-democratic status quo in 1937 precludes a deterministic interpretation. As discussed, institutions such as the revival of the army’s veto power and strengthened judicial independence were endogenous outcomes of political bargaining rather than fixed structural constraints (cf. Negretto Reference Negretto2013 for similar dynamics in Latin America, Opalo Reference Opalo2019 for Africa).
This study adopts a coalition-based approach, emphasizing the balance of power among elites (Waldner and Lust Reference Waldner and Lust2018; Bueno de Mesquita et al. Reference de Mesquita, Bruce, Siverson and Morrow2003). A sufficient number of legislators aligned with pro-army factions, enabling the military to push through anti-democratic policies without formal constitutional amendments. While this case may not be universal, it provides a valuable opportunity to isolate economic and political interests from institutional design (Acemoglu and Robinson Reference Acemoglu and Robinson2006).
Violent repression alone cannot explain these outcomes. Between 1937 and 1941, there were no recorded instances of imprisonment or physical attacks on incumbent legislators. The initial events analyzed here occurred after the last wave of political assassinations in 1936. Only after the army consolidated power did repression intensify. For example, anti-government MP Seigō Nakano was forced to commit suicide in 1943, MP Yukio Ozaki was retroactively jailed for anti-army rhetoric during the 1942 election, and three anti-Tōjō legislators were drafted and sent to war zones (Furukawa Reference Furukawa2001). Before this power shift, democratic institutions shielded legislators from violence, and threats of repression did not appear to deter dissent. Legislators even played a non-trivial role in ousting Prime Minister Tōjō in 1944, underscoring their residual agency.
The findings strongly suggest that weakened legislators were co-opted. Empirical evidence and case studies indicate that these legislators valued the authoritarian legislature as a platform for negotiating with the executive (Reuter and Robertson Reference Reuter and Robertson2013) and were subsequently bought off. Furthermore, the heterogeneity among economic elites casts doubt on ideological shifts such as the rally-around-the-flag effect.
The limited change among procured legislators indicates that their alignment was not driven by coercive resource distribution (Albertus, Fenner, and Slater Reference Albertus, Fenner and Slater2018) or by material dependence on the government (Rosenfeld Reference Rosenfeld2020). Instead, authoritarian forces appear to have targeted weaker and less costly elites for co-optation, while avoiding well-resourced groups whose cooperation was more difficult to secure. This pattern—consistent with Arriola, DeVaro, and Meng’s (Reference Arriola, DeVaro and Meng2021) findings on the co-optation of opposition politicians—highlights how selective collaboration among elites can erode horizontal checks on power (Grillo et al. Reference Grillo, Luo, Nalepa and Prato2024).
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit http://doi.org/10.1017/S0003055426101440.
DATA AVAILABILITY STATEMENT
Research documentation and data that support the findings of this study are openly available at the American Political Science Review Dataverse: https://doi.org/10.7910/DVN/O3VHIX. Limitations on data availability are discussed in the Dataverse.
ACKNOWLEDGMENTS
I would like to thank Paul Pierson, Sarah Anzia, Ernesto Dal Bò, Alison Post, Frederico Finan, Miriam Golden, Dylan Potts, Max Schaub, Damien Bol, Adrián Lucardi, Zeren Li, Shengqiao Lin, Teppei Yamamoto, Masaru Kono, Masato Shizume, Wenkai He, Keiichi Kubo, Nicholas Kuipers, Alexander Sahn, Lukas Leucht, anonymous reviewers, and participants at the seminars in Waseda University, University of Tokyo, European University Institute, Sciences Po, the sessions in Midwest Political Science Association, European Political Science Association, and American Political Science Association. I thank the staff of the National Diet Library of Japan and the Ministry of Defense Archives for facilitating access to archival materials. All remaining errors are my own.
FUNDING STATEMENT
This work was supported by a Grant-in-Aid for Scientific Research (Kakenhi) from the Japan Society for the Promotion of Science (JSPS) under grant numbers 23K1241 and 25K16562.
CONFLICT OF INTEREST
The author declares no ethical issues or conflicts of interest in this research.
ETHICAL STANDARDS
The author affirms this research did not involve human participants.
Open access
Comments
No Comments have been published for this article.