Soft and Sharp Power: An Analysis of China's Asymmetric Warfare

This paper analyzes the bidirectional flow of soft power between the United States and China from 2015 to 2025, measuring soft power through three quantifiable domains: educational exchange, digital platform reach, and cultural tourism. A central finding is the shift from soft power to sharp power as the dominant mode of Chinese international influence, most notably through the Great Firewall of China and the global expansion of platforms such as TikTok. The paper argues that this shift is not the result of superior Chinese cultural appeal but of deliberate institutional asymmetry: China forecloses American influence within its own borders while Chinese-owned infrastructure expands freely through the open American market. The Belt and Road Initiative compounds this pattern at the global level by restructuring political alignments in the Global South in ways that further reduce the reach of American soft power narratives. Together, these mechanisms define a decade-long transformation whose quantitative signature is documented in Table 1 and whose qualitative explanation occupies the body of this paper.

Keywords: soft power, sharp power, Sino-American relations, Great Firewall, Belt and Road Initiative, educational exchange, digital influence, TikTok, gaokao

Introduction

The United States and China represent the two most consequential soft power actors of the modern era, and the dynamics between them have grown increasingly asymmetric over the past decade. Whereas hard power, including military strength, economic coercion, and trade tariffs, has long dominated headlines and formal strategic analysis, the less visible contest of soft power shapes public opinion, cultural affinity, and long-term political alignment in equally consequential ways. This paper focuses on that quieter competition, examining three specific channels through which soft power flows or fails to flow between these two nations: educational exchange, digital media, and cultural tourism.

Soft power, as Nye (2004) defines it, is the capacity to shape the preferences of others through cultural attraction, political values, and the legitimacy of foreign policy, none of which rely on coercion or material incentives. Its contemporary counterpart, sharp power, weaponizes information and institutional control to suppress rival influence while projecting one's own (Nye, 2018). The distinction matters because this paper's central argument is that the decade from 2015 to 2025 witnessed a systematic transition from the former to the latter in China's foreign policy toolkit, and that this transition accounts for the measurable shift in the bilateral soft power balance documented in Table 1.

The thesis of this paper is that the bidirectional flow of soft power between the United States and China shifted decisively over the period 2015 to 2025, moving from approximate balance toward a net advantage for China. This advantage was not produced by superior cultural attraction in the traditional sense that Nye (2004) described when he first developed the concept. It was produced by the systematic deployment of sharp power tools, including the Great Firewall and TikTok, alongside the structural economic leverage of the Belt and Road Initiative. These are not incidental developments. They represent a coherent strategy of asymmetric influence that the Chinese state pursued with deliberate consistency across the entire period under study.

This paper will analyze that shift across the three domains noted above. Each domain provides a distinct and measurable vector for soft power transmission, and together they constitute a multidimensional framework for understanding how influence travels, or fails to travel, across the Pacific. The goal is not to catalog every dimension of Sino-American relations, nor to adjudicate the broader geopolitical competition between the two nations. The goal is to isolate and quantify the mechanisms by which soft power has been redirected, blocked, or amplified within these three specific channels, and to explain why those mechanisms have operated so differently on each side of the relationship.

The paper's geographic and temporal focus is deliberate. Broad global soft power surveys, such as the annual Soft Power 30 (McClory, 2019) and the Global Soft Power Index (Brand Finance, 2025), measure a nation's overall international standing without isolating the bilateral relationship between any two specific countries. This paper addresses that gap by focusing on the directional flows between the United States and China, specifically, across a ten-year window that encompasses both the escalation of trade conflict and the pandemic-era disruptions that reshaped exchange patterns. The bilateral focus allows for a precision of measurement that global indices cannot provide.

The paper proceeds as follows. The Research Strategies section explains the methodological approach and the rationale for the domains and timeframe selected. The Literature Review establishes the theoretical and empirical foundation for the analysis. The Background section provides the political and institutional context in which the decade's shifts occurred. The Method section introduces Table 1 and walks through its quantitative findings in full. Three substantive sections then examine each domain in depth. The paper concludes with a discussion of limitations, a synthesis of findings, and directions for future research.

Research Strategies

This paper employs a mixed-methods approach to analyze the bidirectional flow of soft power between the United States and China from 2015 to 2025. The three domains under analysis — educational exchanges, digital media, and cultural tourism — were selected because each constitutes a distinct and empirically measurable vector of soft power transmission, and because reliable longitudinal data exist for all three across the full study period.

The quantitative backbone of the analysis is the Balance of Exchange ratio, defined as the quotient of inbound-to-outbound human capital and cultural flows between the two nations. Scores are normalized on a zero-to-ten scale to permit comparability across domains, with 2015 serving as the baseline year and 2025 providing the endpoint measurement.

Primary data sources for the quantitative analysis include IIE Open Doors 2025 for international student enrollment figures, the Freedom House Internet Freedom Index for digital access metrics, Pew Research Center (2026) and Medill Spiegel Research Center (2025) for platform reach data, and the U.S. Travel Association (2025) for tourism flows. The U.S.-China Perception Monitor provides supplementary attitudinal data.

Soft power is inherently difficult to quantify because it ultimately depends on public perception, which is intangible, variable across individuals and demographic groups, and resistant to direct measurement. This limitation is common to all quantitative soft power analyses, as the National Academies of Sciences, Engineering, and Medicine (2022) acknowledge, and it does not invalidate the measurement approach. What the quantitative data establishes is the volume and direction of the flows that create opportunities for soft power transmission.

The 2015 to 2025 timeframe was selected for two primary reasons. First, it encompasses the escalation of the U.S.-China trade dispute beginning in 2018. Second, it includes the COVID-19 pandemic, which caused temporary but sharp disruptions to student and tourist flows that must be distinguished from the structural shifts the paper is attempting to identify. The 2025 figures are used as post-pandemic equilibrium markers rather than peak or trough values from 2020 to 2022.

The qualitative component draws on the U.S.-China Perception Monitor, scholarly analysis of the Great Firewall's architecture and effects (Faris and Villeneuve, 2008; King et al., 2013; Gao, 2014), and economic assessments of the Belt and Road Initiative (Lowy Institute, n.d.; Shen and Chan, 2018; UNCTAD, 2019). Each qualitative section exists to answer the question of why the numbers in Table 1 take the shape they do.

A Review of the Literature

The foundational theories of international relations — constructivism, realism, and liberalism — each provide partial tools for analyzing soft power. Nye (2004, 2018) works within a broadly constructivist framework: soft power is non-coercive, attracting rather than compelling, succeeding through legitimacy rather than force. Hard power, as Henry Kissinger (1994, 2011) understood it, operates through domination and depends on the credible threat or exercise of coercive capacity.

Kissinger's analysis of China in On China (2011) argues that China operates from a civilizational logic of patience, indirection, and the gradual accumulation of position rather than confrontation — a framing directly relevant to the Great Firewall, TikTok, and the Belt and Road Initiative, and to why the shift documented in Table 1 accumulated steadily across a decade rather than suddenly.

Acemoglu and Robinson (2012) provide an institutional foundation for understanding how domestic political structures shape the terms of cultural exchange: extractive institutions tend to limit openness and concentrate control, illuminating why China has developed the restrictive digital infrastructure that defines its approach to information.

Jia et al. (2025) offer a concrete example of Chinese soft power operating within the United States. Their analysis of the gaokao, China's national college entrance examination, traces how testing culture rooted in Confucian philosophy has migrated with Chinese immigrant families into American academic communities, embedding an academically rigorous cultural norm bottom-up and without state direction.

General soft power indices — the Soft Power Index (2025), the Soft Power 30 (McClory, 2019), and the Lowy Institute's Asia Power Index (n.d.) — measure a nation's global standing without isolating any specific bilateral relationship. Studies of bidirectional soft power have been conducted for Russia and the United States (Repnikova, 2022) and for China and Middle Eastern partners (Wang and Miao, 2023), but no comparable Sino-American bilateral analysis had been undertaken at the time of writing.

Kim and Diaz (2019) provide the methodological precedent for the Balance of Exchange ratio this paper adapts, demonstrating that directional enrollment figures can serve as a reliable proxy for educational soft power flow. The National Academies of Sciences, Engineering, and Medicine (2022) provide a broader review of measurement methodologies, endorsing the mixed-methods approach this paper follows.

Repnikova (2022) offers a comprehensive treatment of Chinese soft power as a global phenomenon, observing that China's soft power investments have generally been less effective at generating genuine cultural affinity than their proponents hoped — directly relevant to this paper's argument that the bilateral metrics in Table 1 reflect the deployment of sharp power instruments rather than organic cultural growth.

Background

A foundational distinction must be established at the outset. Soft power, as Nye (2004) originally conceived it, is passive and organic. Sharp power, by contrast, is active and state-directed, involving the manipulation of information environments and the coercive projection of influence through institutional means (Nye, 2018). China's trajectory over the past decade represents a systematic transition from the former to the latter.

The political context centers on Xi Jinping's consolidation of power after 2012. Domestically, this manifested in tightening censorship, expanding surveillance infrastructure, and suppressing civil society. Internationally, it manifested in accelerated Great Firewall enforcement, the launch of the Belt and Road Initiative, and increasingly nationalistic foreign policy rhetoric. Lee Kuan Yew observed that China under Xi was recalibrating from Deng Xiaoping's strategic patience toward a more explicit vision of national restoration (Yew, 2013).

The escalation of the U.S.-China trade dispute beginning in 2018 represents the single most significant political disruption to bilateral soft power flows within the study period. Section 301 tariffs on approximately $360 billion in Chinese goods, accompanied by counter-tariffs, technology restrictions, and visa policy changes, created an atmosphere of institutional suspicion affecting educational exchange, tourism, and digital policy simultaneously.

COVID-19 introduced a second major disruption beginning in 2020, sharply curtailing student flows, tourist arrivals, and direct human contact. The methodology normalizes for these anomalies, treating the 2025 data points as representing post-pandemic equilibrium.

On the hard power side, Xi Jinping expanded the People's Liberation Army into new branches, most notably the Rocket Force and the Strategic Support Force, later reorganized into the Information Support Force (Wuthrow, 2021) — context for why soft power competition has grown more adversarial.

The Belt and Road Initiative, launched formally in 2013, deserves particular attention because its relationship to soft power is complex: its direct effects on the bilateral U.S.-China soft power balance are limited, but its indirect effects, constructing durable economic dependencies in regions American soft power once influenced, are considerable.

Method

The analysis proceeds first through quantitative data, for which Table 1 provides the overview, and then through qualitative interpretation organized by domain. The numbers establish what changed; the qualitative analysis explains how and why.

Table 1. Comparison of Bidirectional Soft Power Indicators, 2015–2025

Category / Metric	U.S. Influence in China (2015 → 2025)	Chinese Influence in the U.S. (2015 → 2025)
Education Student enrollment	High → Negligible. U.S. students in China fell from ~15,000 (2012) to ~1,200 (2025), a 92% decline. Normalized score: 7.5 → 0.8. Source: KLCC (2025); IIE historical data	High → Plateau. Chinese enrollment peaked at 372,532 (2019/20), falling to 265,919 (2024/25) — still 22.6% of total U.S. international enrollment. Normalized score: 8.0 → 7.2. Source: IIE Open Doors 2025
Digital / Media Platform access & reach	Nascent → Dominant Adversarial. Google, Meta, YouTube, etc. remain blocked in mainland China. Normalized score: 2.2 → 0.8. Source: Medill Spiegel Research Center (2025); Freedom House	Nascent → Dominant. TikTok grew to 135M+ U.S. users by 2025 (37% of U.S. adults). Normalized score: 0.5 → 9.2. Source: Pew Research Center (2026)
Cultural / Tourism Visitor arrivals	Peaked → Declined Fragmented. Mandatory visas, restricted direct routes, safety advisories. Normalized score: 6.0 → 2.8. Source: News.cn (2026); U.S. Travel Association	Peaked → Significantly Declined. Capped near 3M visitors (2018), declining on visa delays and travel advisories. Normalized score: 6.5 → 3.8. Source: U.S. Travel Association

Note. Scores normalized 0–10. Soft power defined by Nye (2004); sharp power by Nye (2018). Table compiled by the author; all data cited from primary references.

Table 1 provides the quantitative evidence at the center of this paper's argument. Its findings are clear: across all three domains, the bidirectional balance of soft power shifted away from approximate parity in 2015 toward a decisive Chinese advantage by 2025.

In the educational domain, both nations held comparable normalized scores in 2015 (8.0 vs. 7.5). By 2025, American enrollment in China had collapsed to 0.8, a reduction of approximately 92% from peak figures, producing a net flow score of positive 6.4 in favor of Chinese soft power.

The digital domain shows the most dramatic reversal of the three. U.S. digital influence in China fell to 0.8 by 2025 as Great Firewall enforcement intensified, while Chinese influence, propelled primarily by TikTok, rose to 9.2 — the single largest contributor to the aggregate shift.

Tourism presents a different pattern: a broadly symmetric decline in both directions, yielding a smaller net flow of positive 1.0 in China's favor, making this domain less analytically significant for the paper's central argument.

The aggregate picture is stark. In 2015, the U.S. soft power outflow to China across all three domains totaled 15.7 against a Chinese inflow of 15.0. By 2025, the U.S. aggregate had collapsed to 4.4 while the Chinese aggregate rose to 20.2 — a net shift of over 15 points within a single decade.

It is important to interpret these numbers carefully. They do not reflect a sudden growth in the organic attractiveness of Chinese culture, values, or political institutions. What the table documents is the product of deliberate institutional architecture: the Great Firewall suppressing American soft power in China while providing no equivalent protection for the American digital space; TikTok achieving unmatched platform penetration in the United States while remaining structurally inaccessible within China itself; and structural conditions in the educational domain that sustained Chinese educational influence in the U.S. while collapsing American educational presence in China.

Educational and Cultural Exchange

Educational exchange is the oldest and most studied channel of soft power transmission, and it is the domain where the gap between what the data appears to show and what it actually represents is widest. The continued high volume of Chinese students in the United States suggests intense Chinese soft power exposure to American culture; the paradox is that this volume has not produced the reciprocal liberalization traditional soft power theory would predict, and the flow in the reverse direction has effectively been eliminated.

Rooted in Confucian philosophy and formalized through the gaokao, revived by Deng Xiaoping following the Cultural Revolution, the academic culture of Chinese students is organized around rigorous preparation and collective family investment in educational success. Jia, Zhang, and Chen (2025) document how Chinese immigrant families cluster around academically competitive school districts, embedding a testing-oriented culture that was not previously present at comparable intensity — organic transmission of the kind Nye (2004) had in mind.

Chinese enrollment in U.S. institutions fell from a peak of 372,532 in 2019/20 to 265,919 in 2024/25, a reduction of 28%. American enrollment in China, meanwhile, fell from approximately 15,000 in 2012 to roughly 1,200 by 2025, a reduction of approximately 92% — for practical purposes, the elimination of meaningful American educational presence in China.

Several institutional mechanisms produced this asymmetry. The most significant is Optional Practical Training (OPT), which allows international graduates of American universities to remain in the United States for up to three years of post-graduation work experience, prolonging cultural exposure. No equivalent program exists for American students in China. A second mechanism is visa reciprocity friction: Lau (2019) documents how the administrative burden for American students seeking Chinese visas grew substantially more burdensome during the trade dispute era relative to the reverse process. A third is the security climate surrounding Chinese student enrollment in the United States, where exchange increasingly operates as a site of geopolitical friction rather than a medium for mutual understanding.

The conclusion for this domain is directly relevant to the paper's thesis: Chinese soft power flows into the United States through the educational channel at a normalized score of 7.2; American soft power flows into China at 0.8. The mechanisms behind this disparity are institutional (OPT asymmetry, visa friction), geopolitical (the trade dispute, nationalist rhetoric), and cultural (the gaokao effect).

Digital Barriers and the Great Firewall

The digital domain is where this paper's central argument is most clearly demonstrated. If soft power depends on the free encounter and evaluation of a nation's ideas, culture, and political values, then the Great Firewall of China represents the most systematically effective barrier to soft power transmission constructed in the modern era. It does not merely filter; it forecloses.

Through DNS filtering, IP address blocking, and deep packet inspection, the GFW prevents access to the platforms through which American cultural values circulate globally: Google, Facebook, YouTube, Wikipedia, Netflix, Spotify, and most major Western news outlets. Real-time content monitoring compels domestic platforms to practice preemptive self-censorship (King et al., 2013). Faris and Villeneuve (2008) and Gao (2014) confirm this architecture was a deliberate political choice, initiated incrementally beginning in the late 1990s and intensified significantly after 2012.

The consequences for American soft power in China are unambiguous: the normalized score fell from 2.2 in 2015 to 0.8 by 2025, reflecting the near-total exclusion of organic American digital influence from the Chinese market.

While the GFW forecloses American digital influence in China, China simultaneously exploits the absence of equivalent restrictions in the United States. TikTok, owned by ByteDance and subject to Chinese data governance law, grew from its 2016 U.S. launch to more than 135 million American users by 2025, with 37% of American adults using the platform regularly (Pew Research Center, 2026).

A necessary qualification: the content consumed on TikTok by American users is predominantly American in origin. The platform functions, in significant measure, as Chinese infrastructure carrying largely American content. This paper treats TikTok's score as a measure of Chinese platform presence and data infrastructure reach rather than direct evidence of Chinese cultural penetration through content — a distinction the Great Firewall's effects do not share, since its suppression of American soft power in China is unambiguous.

The structural asymmetry of the digital domain is the defining feature of this paper's analysis: one nation operates both an offensive instrument (TikTok) and a defensive barrier (the Great Firewall), while the other operates neither equivalent in either direction. The net digital flow score of 8.4 in China's favor is the quantitative expression of this asymmetry.

Hard Power Developments and the Belt and Road Initiative

The Belt and Road Initiative operates primarily in the Global South and affects the bilateral soft power balance indirectly, through the restructuring of the global environment in which that balance is contested. It does not appear in Table 1's bilateral metrics because its primary effects are not bilateral, but its secondary effects on the conditions under which American soft power operates globally are significant enough to require direct treatment.

The BRI is not a soft power instrument in Nye's (2004) traditional sense. It operates through infrastructure investment, loan disbursements, and the construction of long-term economic dependencies that bind partner nations to Chinese interests through material rather than cultural means — a sharp power instrument that substitutes structural leverage for cultural persuasion.

Over 140 nations have partnered with China under the BRI framework (UNCTAD, 2019; Lowy Institute, n.d.). China thereby constructs durable political alignment without requiring the prerequisite of cultural attraction: partner populations need not be drawn to Chinese culture or values; they need only associate China with the construction of their roads, ports, and power grids.

Shen and Chan (2018) draw the comparison to the Marshall Plan explicitly: the Marshall Plan was accompanied by an explicit ideological framework of democratic institution-building, creating alignment with American values as well as American interests. The BRI generates dependency without an equivalent ideological component — precisely what distinguishes it as sharp power.

Eades (2024) argues that American cultural ubiquity and democratic brand retain a decisive edge in global soft power. Obaidullah and Raihan (2024) counter that in the Global South, the practical benefits of BRI investment carry more immediate political weight than Washington's ideological narratives. This paper takes no categorical position in that debate; what it asserts is the more limited point that the BRI diminishes the effective reach of American soft power in the regions where it operates by making American engagement comparatively less visible and materially relevant.

The Great Firewall is defensive at the domestic level, preventing American soft power from entering China; the BRI is offensive at the global level, expanding Chinese structural influence into regions American soft power once reached through economic engagement. These are two components of a single strategic orientation, with an estimated crossover from soft to sharp power as the dominant mode of Chinese international influence occurring around 2011 — precisely when Xi Jinping consolidated authority.

Discussions and Conclusions

Limitations

This study measures bidirectional soft power through three proxies: student enrollment, platform reach, and tourist arrivals. The most fundamental limitation is the gap between presence and transmission — the Balance of Exchange ratio measures opportunities for soft power transmission, not the actualization of those opportunities, an inherent limitation the entire field faces (National Academies of Sciences, Engineering, and Medicine, 2022).

The TikTok measurement deserves particular attention as the most contested figure in Table 1: it is more precisely a measure of Chinese platform infrastructure than of Chinese cultural content, since the majority of TikTok users and content in the United States are American. The Great Firewall's score of 0.8 does not carry this ambiguity.

The bilateral framework cannot capture indirect transmission pathways — American cultural influence reaching China through India or through South Korean popular culture, itself shaped by American exports, falls entirely outside this study's scope. These omissions mean the paper's findings represent a lower bound on American cultural influence in China rather than a complete accounting of it.

Access to Chinese domestic data represents a further structural limitation: internet usage patterns, content moderation decisions, and TikTok's algorithmic behavior in international markets are not available to external researchers through official channels.

Finally, the conceptual crossover point of approximately 2011 depicted in Figure 2 is an interpolation based on the convergence of several observable developments, not a precise empirical finding. The paper's analytical conclusions do not depend on the precise location of the crossover point; they depend on the much less contestable finding that the trend has moved in the direction the figure depicts.

Conclusion

This paper set out to answer a bounded question: how has the bidirectional flow of soft power between the United States and China changed from 2015 to 2025, measured across educational exchanges, digital media, and cultural tourism? The answer is unambiguous. The balance has shifted decisively in China's favor across all three domains. The United States, which held a slight aggregate soft power advantage in 2015 (15.7 against 15.0), registered a total outflow score of only 4.4 by 2025, compared to a Chinese aggregate of 20.2 — a net shift of approximately 15.8 points over ten years.

That shift is not the product of a sudden growth in the organic attractiveness of Chinese culture, values, or political institutions. If the standard of soft power were cultural attractiveness in Nye's (2004) original sense, the United States would still hold a substantial advantage. The standard this paper has applied measures the actual directional flow of cultural influence in three specific, quantifiable channels — and by that measure, the United States is losing decisively, and the loss is structural rather than cultural.

The explanation lies not in culture but in institutional architecture. The Great Firewall suppresses American soft power in China while providing no equivalent protection for the American digital space. TikTok achieves unmatched platform penetration in the United States while remaining structurally inaccessible to equivalent American market penetration in China. Visa reciprocity imbalances, OPT policy asymmetry, and the political climate of the trade dispute era collectively sustain Chinese educational influence in the United States while collapsing American educational presence in China.

The Belt and Road Initiative compounds this dynamic at the global level. The GFW and the BRI are not separate instruments operating in parallel; they are the defensive and offensive components of a unified sharp power strategy — the former sealing the homeland against foreign influence, the latter extending Chinese structural leverage into the spaces American soft power once occupied.

The United States is not culturally diminished. But the structural conditions that allowed American soft power to flow freely into China have been foreclosed by deliberate Chinese state action, while the structural conditions that allow Chinese influence to flow into the United States remain open by virtue of American cultural and constitutional commitments to a free and open information environment. This asymmetry, and not any failure of American culture itself, is the primary explanation for the imbalance documented in this paper.

Future Research Directions

The role of artificial intelligence in soft power and sharp power competition represents perhaps the most pressing gap. TikTok's recommendation algorithm curates what users see at scale; whether that curation constitutes a form of sharp power independent of any explicit state direction is a question this paper's framework cannot answer with the tools currently available.

The BRICS coalition and its expanding membership deserve analysis as a soft power and structural power framework in their own right, particularly given its recent expansion to include Egypt, Ethiopia, Iran, Saudi Arabia, and the United Arab Emirates. Should China succeed in consolidating BRICS as a coherent alternative economic and political architecture, the bilateral Sino-American framework used here would need to be extended to a genuinely multilateral one.

China's UN peacekeeping deployments, running alongside BRI infrastructure construction in the same regions, create multiple simultaneous channels of influence this paper's bilateral model cannot fully capture.

The indirect transmission pathways noted in the limitations section — American cultural influence reaching China through South Korean, Japanese, or Indian intermediaries — represent a significant methodological gap future bilateral analyses should attempt to address.

Finally, the longitudinal scope of this analysis should be extended both backward, to the 1995–2015 period encompassing China's WTO accession and the expansion of Confucius Institutes, and forward, to assess whether the structural asymmetries documented here are permanent features of the bilateral relationship or subject to policy intervention, technological disruption, or geopolitical reconfiguration.