1. Introduction
One of the long-standing motivations for foreign aid is to capture the hearts and minds of the public of recipient countries. The US and the Soviet Union used foreign aid to influence and dominate other countries with their political values during the Cold War. More recently, China's growing power on the international stage and the influence of the country's aggressive aid on recipient citizens' attitudes toward China or the US have been keenly researched (Blair et al., Reference Blair, Marty and Roessler2022; Wellner et al., Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022). Empirical studies on whether foreign aid indeed improves the donor country's image among the recipient country's public are inconclusive. Some find a positive effect (Goldsmith et al., Reference Goldsmith, Horiuchi and Wood2014; Dietrich et al., Reference Dietrich, Mahmud and Winters2018; Mattingly and Sundquist, Reference Mattingly and Sundquist2022), while others suggest a null or negative effect (Zürcher, Reference Zürcher2012; Tokdemir, Reference Tokdemir2017; Green-Riley, Reference Green-Riley2020).
This study departs from the existing studies on foreign aid and national image that have focused on the donor country's image among the recipient country's public, what we call a first-order effect, and explores whether foreign aid might have the impact of improving the donor country's image among the third country's public. Our focus on this second-order effect of foreign aid speaks to the growing scholarly and practitioners' interest in the deployment of soft power resources in public diplomacy. As scholars of international relations have largely come to the consensus that the public's general attitudes toward a foreign country can influence the foreign policy toward that country (Goldsmith and Horiuchi, Reference Goldsmith and Horiuchi2012; Dietrich et al., Reference Dietrich, Mahmud and Winters2018), many countries are implementing public diplomacy policies.Footnote 1 Particularly, ‘soft power,’ the ability to obtain what one wants through attraction rather than coercion or payments (Nye, Reference Nye2008), is considered to be a significant component of public diplomacy (Gilboa, Reference Gilboa2008). While foreign aid in the form of economic payments can be considered as the use of hard power by the donor country in its relationship with the recipient country, the act of providing aid to a developing country can project the donor's soft power toward a third country.
Hence, our main research question is as follows: Does foreign aid influence how non-recipient countries' public think about donor countries? As discussed in the Theoretical Framework section, not all foreign aid activities are expected to bring out the desired second-order effect. We discuss four general features of aid that are most likely to improve the donor country's image among the public of a third country: it is visible, has no political strings attached, is localized, and deals with the global public good. We then identify cultural aid to restore the United Nations Educational, Scientific and Cultural Organization (UNESCO) world heritage sites as the most likely type of foreign aid to meet the four conditions.
We empirically investigate whether this type of aid can generate a second-order effect using a survey experiment in Australia. Australia is a highly developed, non-aid-receiving country whose citizens do not expect any economic payment from donor countries. We thus believe Australia is an appropriate choice as a third country to assess the soft power of foreign aid. As donor countries, we chose China and South Korea (hereafter Korea). This is in light of existing studies suggesting that a hostile bilateral relationship undermines the effectiveness of public diplomacy (Entnam, Reference Entnam2008; Sheafer and Gabay, Reference Sheafer and Gabay2009; Sheafer et al., Reference Sheafer, Shenhav, Takens and Van Atteveldt2014). While Australia has maintained an amicable relationship with Korea, its relationship with China has been characterized by a severe trade war which affected the public in both countries in the terms of security, economic, and energy. The Australian respondents are randomly exposed to video treatment on foreign aid that is provided by either China or Korea to restore Cambodia's Angkor monuments. Other than information about the donors, these videos are nearly the same in terms of the images, narration, length, and background music.
We find that information on Korea's aid to Cambodia significantly improves Australians' feelings toward Korea. Additionally, information on Korea's aid has a weak but still positive effect on eliciting the willingness of Australians to support their government's cooperation with the Korean government. Neither type of effect, however, is observed for information about China's aid to Cambodia. Our finding provides first-cut empirical evidence that certain types of foreign aid can be utilized as a soft power resource in public diplomacy, particularly toward non-recipient countries whose bilateral relationship with the donor country is relatively amicable.
2. Foreign aid and attitude toward the donor country
In this section, we begin by reviewing the literature on how foreign aid shapes the donor country's image in the recipient country, that is, the first-order effect of foreign aid. While empirical findings from this strand of the literature remain inconclusive, a systemic review of the studies yields two important takeaways.
First, who provides foreign aid appears to matter. Blair et al. (Reference Blair, Marty and Roessler2022) find that, while US aid to African countries increased support for the US, Chinese aid to African countries reduced beneficiaries' support for China. The negative perception among recipients that the Chinese aid regime prioritizes ‘speed and low cost at the expense of quality,’ is one tentative explanation that the authors provide for this difference. Similarly, Eichenauer et al. (Reference Eichenauer, Fuchs and Brückner2018: 20)'s analysis of US and Chinese aid to Latin American countries finds that ‘Chinese aid has not (yet) helped increase its public image abroad, while American aid does so for the United States.’ In addition to its low quality, the authors also suggest that Chinese aid is more prone to corruption and misappropriation and less likely to reach the general public.
Second and relatedly, the type of aid matters. Goldsmith et al. (Reference Goldsmith, Horiuchi and Wood2014) suggest that foreign aid that is effective, visible, and sustained could foster a positive first-order effect; for instance, a US health aid program that meets these conditions significantly improved the perception of US leadership within the recipient countries. Dietrich et al. (Reference Dietrich, Mahmud and Winters2018) also find that US health clinics projects in Bangladeshi improved citizens' general attitudes toward the US. More recently, Mattingly and Sundquist (Reference Mattingly and Sundquist2022) find that when Chinese diplomats touted aid that donating safety protection supplies to Indian citizens via twitter, it improved the citizens' perception of the Chinese government, people, and countermeasure to the COVID-19 pandemic. In contrast, less-tangible types of Chinese aid, such as its education and cultural aid provided via Confucius institutes, have been found to be ineffective in fostering positive perceptions of China (Green-Riley, Reference Green-Riley2020).
To our knowledge, few studies have investigated how foreign aid shapes the image of the donor country in non-recipient countries. One exception is Zielińska (Reference Zielińska2016), who suggests that foreign aid could build up a positive image beyond the recipient countries when coupled with other instruments such as promotional campaigns. Although the author discusses the significance of and possibilities associated with the second-order effect of foreign aid, she does not provide any empirical evidence for this. A recent study by Wellner et al. (Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022) attempted to estimate the second-order effect of aid. Using survey data covering a large number of developing countries, including those that do not receive Chinese aid, the authors find that Chinese aid improved public opinion about China within African countries, potential swing states in the United Nations General Assembly, and countries that already had a high level of public support for the Chinese government (Wellner et al., Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022). Incerti et al.'s (Reference Incerti, Ju, Mattingly, Moreshead, Rosenbluth, Tanaka and Yamagishi2022) working paper is the only study we are aware of that empirically evaluates the second-order effect of foreign aid in developed countries. They investigated how Chinese and Japanese foreign aid was received by the public in Korea and Malaysia by providing them with general information about bilateral aid from China and Japan. The authors found that in Malaysia, where bilateral relations with both China and Japan were relatively depoliticized, promoting foreign aid produced positive attitudes. In Korea, however, the effects were more ambiguous.
It is noteworthy that both Incerti et al. (Reference Incerti, Ju, Mattingly, Moreshead, Rosenbluth, Tanaka and Yamagishi2022) and Wellner et al. (Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022) find a positive second-order effect of foreign aid when the donor enjoys an amicable relationship with the third country. As we discuss in the next section, this finding is in line with existing studies on public diplomacy that suggest prior relationships to be an important condition for more effective public diplomacy (Entnam, Reference Entnam2008; Sheafer and Gabay, Reference Sheafer and Gabay2009; Sheafer et al., Reference Sheafer, Shenhav, Takens and Van Atteveldt2014).
3. Theoretical framework
We add to the nascent literature on the second-order effect of foreign aid (Zielińska, Reference Zielińska2016; Incerti et al., Reference Incerti, Ju, Mattingly, Moreshead, Rosenbluth, Tanaka and Yamagishi2022; Wellner et al., Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022) by focusing on what makes a positive effect in a third county more likely. We first identify four general features of an aid project that are likely to generate a positive second-order effect. We then discuss how the existing bilateral relationship with the third country might condition the effect.
3.1 Aid project-varying effect
3.1.1 Visibility
To study public attitudes toward a donor country, the first consideration is how the public gains information about the donor country's aid projects. Under what circumstances can the general public, in particular people outside the recipient country, become aware of foreign aid projects? Direct exposure occurs when the third-party audience travels to a recipient country and encounters a visible aid project site by chance. The third-party audience may also learn of certain aid projects indirectly via the active promotion of the donor country using media coverage of project ceremonies and achievements (Dietrich et al., Reference Dietrich, Mahmud and Winters2018; Wellner et al., Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022). Aid projects involving visible sites and tangible results, such as the funding of a new hospital or school, are easier to promote than less-visible aid in areas such as social cohesion and capacity building. Second-order audiences themselves can become aware of and appreciate these visible projects without much cognitive effort. Less-visible aid projects, in contrast, require more effort and creativity to promote. As Vollmer (Reference Vollmer2012: 64) point out, ‘if the results [of a particular aid project] are not visible as such (which can easily be the case in such areas as gender or fostering social cohesion, which are, by definition, not as visible as, say, the funding of a new building or street), they should be made visible, through creative attempts at visualising data and information and improved communication.’ We thus suggest that the visibility of an aid project is an important feature that facilitates the second-order effect.
3.1.2 Aid with no political strings attached
Aid can often be politicized when donor countries pressure recipient countries to implement some policies or adopt certain ideologies. Some scholars even maintain that foreign aid has always been political, and that it should not be surprising when development aid is politicized (Novelli, Reference Novelli2010). Foreign aid has long been used in conflict situations to drive politicized agendas (Boutros-Ghali, Reference Boutros-Ghali1992). For example, during the post-World War II period, aid was allocated based on political alliances rather than humanitarian needs (Lundborg, Reference Lundborg1998; Wang Reference Wang1999; Novelli, Reference Novelli2010); education development aid in Iraq and Afghanistan provided by Western countries after 9/11 was considered as a political tool of the occupying force in the context of the battle between the West and Islamic terrorism (McClure, Reference McClure2009; Novelli, Reference Novelli2010).
Even in non-conflict situations, foreign aid in terms of the dissemination of ideas, allocation of humanitarian resources, and implementation of relief projects is provided through subtle power processes (Hilhorst and Jansen, Reference Hilhorst and Jansen2013). Foreign aid can be used as a bribe to obtain the allegiance of non-permanent members of the UN Security Council (Kuziemko and Werker, Reference Kuziemko and Werker2006; Vreeland and Dreher, Reference Vreeland and Dreher2014). As such, the public in a third country may view bilateral foreign aid as a pursuit of the donor's political interest rather than pure generosity for the recipient's citizens (Goldsmith et al., Reference Goldsmith, Horiuchi and Wood2014; Tokdemir, Reference Tokdemir2017). We thus suggest that aid that seems the least politicized, for instance, by being channeled through an authoritative multilateral institution, would have more potential to generate a second-order effect.
3.1.3 Localization
In recent years, the concept of ‘localization’ has become increasingly prominent in international aid to enhance the ownership of recipient countries and avoid unilateral progress by donor countries (Bonacker et al., Reference Bonacker, von Heusinger and Zimmer2017; Hickey, Reference Hickey2017; Sundberg, Reference Sundberg2019).Footnote 2 If an aid project is performed without respect for the culture and environment of the recipient country, it can evoke ideas of imperialism and colonization among not only the recipient country's public but also a third country's public. In this sense, localization of aid is an important condition for generating the second-order effect. We thus expect the second-order effect to be generated more effectively when donor countries can signal their cultural sensitivity and respect for a recipient country by, for instance, listening to the recipient country's demands, allowing locals' participation in every process, using local materials and, finally, handing over the project to the locals (Van Brabant and Patel, Reference Van Brabant and Patel2018).
3.1.4 Global public goods
To maximize the first-order effect (i.e., the effect on a recipient country's public), targeted aid with a clearly defined set of recipients can be more effective (Goldsmith et al., Reference Goldsmith, Horiuchi and Wood2014). To generate the second-order effect (i.e., the effect on a third country's public), however, we expect foreign aid projects whose benefits reach beyond the narrow population of the recipient to be more effective. What are such projects? We zoom in on those projects that contribute to the provision and maintenance of global public goods. According to Kaul et al. (Reference Kaul, Grunberg and Stern1999), global public goods are public goods with strong qualities of broad usage and universality in terms of countries, people, and generations. Development projects can contribute to the provision of global public goods such as knowledge generation and dissemination, eradication of communicable diseases, global commons, free and open trade systems, and international financial stability (Speight, Reference Speight2002). These projects create positive externalities and thereby expand the range of beneficiaries. We expect such projects to be appreciated by the public in third-party countries and thus have potential to generate a significant second-order effect.
Thus far, we have identified several conditions of foreign aid that we expect to be ideal for the generation of second-order effects. Not all aid projects are likely to meet all these conditions. For example, although aid for the stability of international finance or aid for trade systems contributes to the provision of global public goods, it is difficult to showcase it since its elitist nature means that lay individuals cannot easily understand it. Beyond the liberal world, the promotion of a liberal economic order through this type of aid can also be considered quite politicized. Although aid for national infrastructure projects, such as airports, roads, and ports, is quite visible, it is likely to be less appreciated by a third country's public. Thus, aid that meets some of our conditions often fails to satisfy the others.
In this study, we focus on aid to restore UNESCO's world heritage sites as among the most likely cases to meet all our conditions. Specifically, aid projects for the restoration, renovation, or conservation of world heritage sites usually satisfy the conditions. Cultural and natural heritage sites are visible because people from many countries visit these sites for enjoyment. A famous heritage site being transformed from a shattered place to the magnificently recovered one is noticeable for most people without much cognitive effort. This type of aid is hardly ever politicized because it neither promotes ideologies nor imposes policies. The fact that an authoritative multilateral organization, UNESCO, manages the list of world heritage sites further mitigates the politicization of the projects. Aid to world heritage sites is also often highly localized because recipient countries' experts and communities, who have in-depth knowledge of local culture and history, work together on these sites. In addition, the projects themselves are based on an appreciation of the recipient countries' cultures and traditions. Last, world heritage sites represent global public goods that diffuse positive externalities to national and international visitors and non-visitors (Serageldin, Reference Serageldin, Kaul, Grunberg and Stern1999; Bodansky, Reference Bodansky2012). Therefore, aid projects to world heritage sites are likely to generate a positive second-order effect. Thus, our first hypothesis is as follows:
Hypothesis 1: Foreign aid to restore UNESCO's world heritage sites will improve the image of donor nations among the public of third-party countries.
3.2 Prior relationship-varying effect
The previous subsection discussed four general features of an aid project that we expect to generate a positive second-order effect. We then identified the use of aid to restore UNESCO's world heritage sites as one of the most likely cases to meet these conditions. Not all countries, however, are equally capable of utilizing foreign aid to world heritage sites to improve their image in third-party countries. This is where the literature on public diplomacy effectiveness becomes relevant. A number of empirical studies find that public diplomacy activities tend to be more effective when the relationship with the target government and the target public is already amicable (Nye, Reference Nye2004; Entnam, Reference Entnam2008; Sheafer and Gabay, Reference Sheafer and Gabay2009; Sheafer et al., Reference Sheafer, Shenhav, Takens and Van Atteveldt2014). In contrast, when diplomatic tensions with target countries arise, public diplomacy can be ineffective or can even backfire (Goldsmith et al., Reference Goldsmith, Horiuchi and Wood2014; Mattingly and Sundquist, Reference Mattingly and Sundquist2022). This effect of prior relationships has also been reported by recent studies on the second-order effect of foreign aid (Incerti et al., Reference Incerti, Ju, Mattingly, Moreshead, Rosenbluth, Tanaka and Yamagishi2022; Wellner et al., Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022).
What explains these findings? Due to negativity bias in the media (Lengauer et al., Reference Lengauer, Esser and Berganza2012; Soroka and McAdams, Reference Soroka and McAdams2015), diplomatic tensions tend to have higher public salience than amicable relations. The literature on persuasive communication suggests that people are more reliant on simple cues when knowledge is low or inaccessible, but as general knowledge about a topic increases, people think more critically about new information (Wood and Kallgren, Reference Wood and Kallgren1988; Crowder-Meyer et al., Reference Crowder-Meyer, Gadarian and Trounstine2020). In particular, when new information goes against their prior knowledge, people tend to downplay it and may even refuse to accept it to prevent cognitive dissonance, which is the uneasy feeling of holding inconsistent beliefs (Edwards and Smith, Reference Edwards and Smith1996; Svoboda, Reference Svoboda2017). Hence, generating the positive second-order effect of foreign aid would be easier when the bilateral relations with the target country has been amicable and of low public salience. In contrast, when the target public has already been exposed to negative news about the donor country and holds negative attitudes toward it, it would be more difficult to change the target public's beliefs. Foreign aid that is employed as a public diplomacy tool may be ineffective or may even backfire when the efforts to create positive feelings are resisted by the public in a third country.
Hypothesis 2: In an unfriendly relationship (unfavorable image or attitude toward the donor country) the second-order effect will be weakened.
4. Research design
4.1 Selection of the aid project
As a foreign aid project that is highly visible, localized, has no political strings attached, and deals with the global public good, we focus on a world heritage site restoration project in Cambodia called the ‘Restoration and Construction of Angkor Monument Project.’ Angkor is an archaeological site that contains several historic temples. It became a main source of national income after it was named a UNESCO world heritage site in 1992. However, the temples in Angkor were seriously damaged by heavy rains, looting, and a lack of maintenance. To restore the site, the Cambodian government requested international assistance, and Angkor was inscribed on the List of World Heritage in Danger in the same year. Seventeen countries, including Korea, Japan, France, and China, helped restore and sustainably protect this heritage site. The case of Angkor was chosen because it was a successful case of the restoration of a world heritage site; through the efforts of aid projects, Angkor was removed from the List of World Heritage in Danger in 2004.
4.2 Selection of the third-party country and donor countries
Our empirical analysis employs data from an original survey conducted online using a sample of 872 Australians in October 2021.Footnote 3 Australia is a highly advanced country that does not receive any foreign aid, which makes it an appropriate third country to assess the soft power of foreign aid. The positive change in the image of the donor country, if any, can be attributed to Australians' appreciation of the donor country's soft power (such as its generosity, cultural values, and leadership), rather than their anticipation of economic benefits.
Our survey includes an experimental section in which respondents are randomly assigned to a video treatment about the Angkor restoration aid project of either China or Korea. We chose to deliver aid information about these two donor countries considering the difference in their bilateral relations with Australia, which is important for testing our second hypothesis (H2). According to the Australian Government Department of Foreign Affairs and Trade (2021), Australia and the Republic of Korea have had an amicable relationship for a long time. The declaration of a comprehensive strategic partnership between the two countries at the G7 meeting in June 2021 reaffirms that their relationship is strong. In the dialogue between foreign and defense ministers of the two countries, they agreed to enhance defense cooperation through increased joint training and exercises and develop a higher level of bilateral interoperability between their navies. Hence the relationship between Australia and Korea was smooth when the respondents take the survey and until now. In contrast, China has been in a trade war with Australia starting from April 2020. Through the trade war, China has forbidden the import of beef, wine, cotton, wood, copper, lobster, etc. Australia lodged a formal complaint with the WTO over China's imposition of anti-dumping duties on Australian products' exports. The trade dispute was lasted over a year and expanded to diplomatic and security problems.
Reflecting these divergent bilateral relationships, according to the annual Lowy Institute Poll, Australians have consistently shown warmer attitudes toward Korea than toward China over the past decade. In recent polls, Korea scored 62 (2019), 57 (2020), 61 (2021), and 63 (2022), whereas the rate for China was much lower and continuously fell from 49 (2019) to 39 (2020), 32 (2021), and 33 (2022). This shows that the image of Korea in the eyes of Australians is substantially more positive than that of China.Footnote 4 Thus, in light of H2, we expect a new piece of positive information about Korea's aid project to have a stronger belief-changing effect than a similar piece of information about China's aid project.
4.3 Survey design
As mentioned above, our survey in Australia includes an experimental section in which the respondents are randomly assigned to different video treatments (see Table 1). One of the clips is about an aid project funded by Korea called the ‘Conservation and Restoration Project for the Preah Pithu temple group in Angkor Complex.’ Another clip is about a Chinese-funded project, the ‘Restoration of Chau Say Tevoda temple of Angkor Wat.’ Last, there is a neutral clip that provides nearly the same general information on the Angkor site that was provided in the treatment videos but with just longer sentences and without any information on a donor-specific aid project.
The control and treatment groups
We tried our best to make the three videos as similar as possible and preclude any confounders for the treatment effect. All three clips were of 86 seconds and used the same narrator's voice, background music, and general information about Angkor sites. We used the same Angkor monument images from a National Geographic documentary (In Cambodia, a City of Towering Temples in the Forest). Please see the Appendix for the full transcripts.
To ensure that the respondents watched the video, we added a setting that forced them to spend a certain amount of time viewing the video before moving to the next section.
4.4 Outcome variables (feeling thermometer and willingness to cooperate)
To understand the attitudes toward the donor nations, we employed a ‘feeling thermometer’ measure, a scale ranging from 0 to 100, with 0 indicating ‘very cold’ and 100 indicating ‘very warm’ feelings toward a subject. We asked the respondents to evaluate their feelings toward Korea and China. In addition, we measured the willingness to cooperate with Korea and China. The question reads as follows: ‘Do you agree that the Australian government should cooperate more with the following countries?’ The answers were recorded on a five-point scale from 1 (strongly disagree) to 5 (strongly agree), with 3 serving as a neutral point. To not explicitly reveal the purpose of the study, we collected the feeling thermometer and willingness to cooperate measures for four additional Asian countries (Cambodia, Japan, Thailand, and Vietnam) in a randomized order.
4.5 Control variables
For more precise estimation of the treatment effect, we controlled for two basic demographic variables: Female (1 if male, 2 if female) and Age: (1) 18–24 years old, (2) 25–34 years old, (3) 35–44 years old, (4) 45–54 years old, (5) 55–64 years old, and (6) 65 years or older. We also controlled for variables that capture familiarity with a donor country. First, Korea Visit and China Visit are dummy variables regarding whether an individual has visited the donor country (Korea or China). We recorded (1) as visited and (2) as never visited. The other variables for capturing familiarity with a donor country are Korean Friend and Chinese Friend. We recorded (1) if a respondent had a friend from the country and (2) if they did not. Individuals with a higher familiarity with a donor country may show more positive attitudes toward the donor country.
5. Results
Table 2 presents the results of the two sets of regressions for each dependent variable, one with the basic control variables (Age and Female) and another with more control variables (Korea Visit, China Visit, Korean Friend, and Chinese Friend).Footnote 5 The first four models use the treatment of watching a video on Korea's aid as the main independent variable (Korean Aid). The next four models use the treatment of watching a video on China's aid as the main independent variable (Chinese Aid).
Main models
Notes: +P < 0.1; *P < 0.05; **P < 0.01; ***P < 0.001; standard errors in parentheses.
In models 1–2, the significant and positive coefficient estimate for Korean Aid suggests that showing a video on the Korean aid project significantly improves the participants' feelings toward Korea. Figure 1, based on model 2, presents the feeling thermometers in respect of Korea in each of the three groups. The solid lines indicate a 90% confidence interval. The figure reveals that the group who watched the video on Korea's aid project exhibited the warmest feelings toward Korea. In models 3 and 4, we further examine whether the treatment influences the willingness to cooperate with the Korean government. Although the relationship is weaker (P < 0.1), the coefficient estimate for Korean Aid in model 4 is positive.
Feeling thermometer for Korea.
Note: The predicted values are based on model 2.
Figure 2 presents the effect from model 4. The Y-axis indicates the predicted probability of agreeing that the Australian government should cooperate more with Korea (4 and 5 on a scale of 1–5) with 90% confidence intervals. While the high increase of the feeling thermometer is statistically significant, the treatment effect on the willingness to cooperate is modest. This is consistent with the existing literature that suggests that the feeling thermometer as an emotional indicator can easily be swayed by external shocks (Li, Reference Li2021), but that it is more difficult to change opinions. Nevertheless, the difference made by the video of less than 90 seconds was significant. The findings together lend support to Hypothesis 1.
Willingness to cooperate with the Korean government.
Note: The predicted values are based on model 4.
By contrast, as shown in models 5–8, we do not find any evidence that watching the video on Chinese aid influences Australian citizens' feelings toward China. This is consistent with Hypothesis 2. Despite the similarity of the content of the video clips, no second-order effect seems to be generated for Chinese aid. Please see the Appendix for the figures that visualize this null finding.
In addition to the regression analysis, to qualitatively assess the causal mechanism, we analyze responses to two open-ended questions that were asked at the end of the survey: ‘Why did you rate the feeling thermometer for China [Korea] with that grade?’ and ‘Why do you think/Why do you not think the Australian government should cooperate with China [Korea]?’ Examining the responses to the questions regarding China, we could infer why the second-order effect did not occur. There were answers suggesting suspicion about China's intentions in the treatment group that watched the video on Chinese aid: ‘China is trying to expand its control over many of the Pacific nations by supplying funds etc. to develop the country.’ ‘I am not convinced about China's intentions.’ We propose that such attitudes led the respondents to doubt the value of the aid project. Responses such as, ‘I think they are doing more harm than good to other countries,’ and ‘Because this country is not very nice with other surrounding countries in Asia’ indicate that the information about the aid project was not sufficient to change their prior image of China.
To the contrary, in the same questions about Korea, the responses include ‘They are doing a wonderful thing for another country,’ and ‘They helped restore the ruined city in Cambodia.’ Other responses were: ‘It seems like it is a country doing good things for the world,’ ‘Korea is a modern and advanced country, it contributes positively to the world and its neighbors,’ and ‘They are working with Cambodia to restore UNESCO sites.’ We identify that some people not only view the Korean aid as helping a neighbor country, but also as doing a good thing for the whole world since it is a UNESCO World Heritage Site. This shows that people value global public goods, consistent with the theoretical expectations.
5.1 Interaction models: Which individual-level conditions intensify the second-order effect?
We zoom in on the effect of Korean Aid and further investigate how the second-order effect might be intensified by individual-level characteristics. In light of the studies that suggest familiarity gained by prior interactions and exposures breeds attraction (Ebbesen et al., Reference Ebbesen, Kjos and Konečni1976; Peskin and Newell, Reference Peskin and Newell2004; Reis et al., Reference Reis, Maniaci, Caprariello, Eastwick and Finkel2011), we first examine whether familiarity with a donor country may magnify the positive second-order effect. For this, we introduce an interaction term for Korean Aid and proxies for familiarity with Korea. Respondents who have traveled to Korea (Korea Visit) or have a Korean friend are expected to be more familiar with Korea than those who have never been to Korea or who have no Korean friend. We expect our information treatment to have a more positive effect among those with higher familiarity.
Our results are reported in Table 3. Model 9 investigates Korean Aid × Korea Visit, while model 10 investigates Korean Aid × Korean Friend. In both models, we do not find any evidence that an Australian who is more familiar with Korea reacts more positively to our treatment. This may be because the baseline attitudes toward Korea among those familiar with Korea were already quite positive. Indeed, our data show that the average feeling thermometer for those who have been to Korea is 80.50 compared to only 64.09 for those who have never been to Korea, and 81.36 for those who have a Korean friend compared to only 62.37 for those who do not.
Interaction models
Notes: +P < 0.1; *P < 0.05; **P < 0.01; ***P < 0.001; standard errors in parentheses.
Last, we explore the interaction between our treatment and the respondents' appreciation of the heritage site. Model 11 investigates Korean Aid × Value. It includes a variable that captures how much respondents value the Angkor site: ‘Do you consider the Angkor site a valuable world heritage site?’ The scale ranges from (1) Not valuable at all to (5) Very valuable.Footnote 6 We expect the second-order effect to be generated primarily among those who value the site. Most respondents in our sample (82.48%) indeed believe the site is valuable (4) (28.82%) or very valuable (5) (53.66%). The result from model 11, which is visualized in Figure 3, supports our expectation. It is among those Australians who appreciate the value of the cultural heritage site that the image of Korea improves (the interaction term is positive and significant at the 0.1 level).Footnote 7 Overall, our findings provide empirical evidence that foreign aid as soft power spreads beyond a recipient country.
Korean aid interaction with perceived value of Angkor.
Note: The predicted values are based on model 11.
6. Discussion
This study adds to the nascent literature that explores whether foreign aid can influence the donor country's image among a third-party country's public (Zielińska, Reference Zielińska2016; Incerti et al., Reference Incerti, Ju, Mattingly, Moreshead, Rosenbluth, Tanaka and Yamagishi2022; Wellner et al., Reference Wellner, Dreher, Fuchs, Fuchs, Parks and Strange2022), what we refer to as the second-order effect of foreign aid. In doing so, this study makes several theoretical contributions. First, we identify four general features of aid that are likely to generate a positive second-order effect: it is visible, has no political strings attached, is localized, and serves the global public good. While the empirical component of this study focuses on a UNESCO world heritage site restoration project, our theoretical expectation is that other types of aid that meet these four conditions would similarly generate positive second-order effects.
At the same time, our study reaffirms the key insight of the literature regarding the importance of preexisting relationships in conditioning the effectiveness of public diplomacy (Entnam, Reference Entnam2008; Sheafer and Gabay, Reference Sheafer and Gabay2009; Sheafer et al., Reference Sheafer, Shenhav, Takens and Van Atteveldt2014; Incerti et al., Reference Incerti, Ju, Mattingly, Moreshead, Rosenbluth, Tanaka and Yamagishi2022). While information on Korea's aid improves the image of Korea and the willingness to cooperate with the Korean government among Australians, no such effect is observed in the case of similar aid by China whose relations with Australia have been strained in multiple domains.
Our study also has a number of limitations that suggest interesting future research directions. First, while we discuss four general features of aid projects that have the strong potential to generate a second-order effect, our empirical research design focusing on one case does not allow us to disentangle how each of these four features contributes to the second-order effect. Future studies can utilize a more rigorous research design to investigate the second-order effect potential of different aid projects that vary systematically across the four features. Future research can also vary the profile of the recipient country to test the generalizability of our findings.
Another interesting avenue for future research is to examine the role of various communication strategies in tapping into foreign aid's soft power potential among a broad audience. Effective and creative messaging might allow donors to generate a second-order effect even with aid projects that do not necessarily meet all of the four features discussed in this study. Lastly, our research invites future empirical research regarding how the positive second-order effect of foreign aid might be used, in turn, to sustain and increase domestic public support for foreign aid within donor countries. Experimental research can examine whether information about the second-order effect of foreign aid (i.e., improved national image in third-party countries) can encourage aid skeptics to become more supportive of their government's foreign aid policy and of increasing the foreign aid budget.
7. Conclusion
Can foreign aid be used to capture the hearts and minds of the foreign public? Our study is one of the first attempts to take this inquiry beyond donor–recipient relations and explore whether foreign aid can influence the donor country's image among the public of a third-party country. Our finding suggests that the power of foreign aid is not limited to the recipient country; it may extend further than we think.
We believe our findings have important policy implications for practitioners in donor countries and those in public diplomacy. Governments can utilize the soft power of foreign aid in public diplomacy to improve their country's image abroad and foster cooperation. For this, they should more effectively promote their aid projects with third-party countries' public in mind, especially regarding aid that has the four features we identified in this study. Our results also imply that, to maximize the second-order effect, governments can strategically target the publics of those third-party countries with which they have relatively amicable bilateral relations. Within the donor country, our survey's experimental design can be used by aid agencies and project managers to evaluate the soft power and public diplomacy-related outcomes of aid projects.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1468109922000299 and https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/DSWBH0.
Conflict of interest
None.
Open access
