Cultural change and the migration choice

Abstract

Cultural differences play an important role in shaping migration patterns. The conventional proxies for cross country cultural differences, such as common language; ethnicity; genetic traits; or religion, implicitly assume that cultural proximity between two countries is constant over time and symmetric. This is far from realistic. This paper proposes a gravity model for international migration which explicitly allows for the time varying and asymmetric dimensions of cultural proximity. In accordance with Disdier, Tai, Fontagné, Mayer (Rev World Econ, 145(4):575–595, 2010) we assume that the evolution of bilateral cultural affinity over time is reflected in the intensity of bilateral trade in cultural goods. The empirical framework includes a comprehensive set of high dimensional fixed effects which enable identification of the impact of cultural proximity on migration over and beyond the effect of pre-existing cultural and historical ties. The results are robust across different econometric techniques and suggest that positive changes in cultural relationships over time foster bilateral migration.

Introduction

Harris and Todaro (1970) interprets migration flows in terms of the wage differential between sending and destination countries and the associated cost of the journey. This interpretation has been seen as insufficient in explaining migration patterns. Indeed, the presence of fairly small migration flows—both within and between countries—has shifted the focus of the literature away from economic factors, despite very strong economic drivers such as unemployment and wage differentials (see Barro & Sala-i-Martin, 1995). More attention is now given to non-pecuniary determinants of migration decisions, such as cultural relationships.

Earlier empirical research has shown that measures of cultural proximity—e.g. bilateral linguistic, religious and genetic distance as well as colonial links—are often more important determinants of migration patterns than traditional economic variables.¹ The work of Belot and Ederveen (2012), in particular, provides sound empirical evidence on the central role of cultural distance in shaping migration patterns. They analysed the impact of several dimensions of cross-country cultural barriers using a set of indicators describing bilateral religious and linguistic distance. They also included survey-based composite measures of cultural distance, such as the Hofstede or the Inglehart and Baker indexes, with all of them fostering bilateral emigration rates. Similar proxies of cultural proximity have been found by Belot and Hatton (2012) to be more important determinants of educational selectivity in immigration than wage incentives or a selective immigration policy.

All of these measures, however, have been challenged in their capacity to effectively capture some important dimensions of cultural relationships (see Shenkar 2001; Li et al. 2017; Felbermayr & Toubal 2010; Tung & Verbeke 2010), which we understand as being particularly relevant for international migration. More specifically, they are unlikely to be able to fully capture a broader notion of cultural proximity which hinges on the acknowledgement that cultural relationships are subject to variation over time.² Measured at a single point in time, existing measures of cultural proximity are considered to be constant. However, this is not realistic as culture measured when the decision to migrate was made may have changed by the time that culture is measured. Recent macro indexes of cultural distance based on Hofstede’s cultural dimensions do not account for the time dimension (see Kaasa et al., 2016) and they are mostly confined to European countries that are more culturally alike.³ The assumption of stability is particularly unrealistic when we consider the last two decades which have been characterized by citizens’ growing exposure to foreign cultures through cross‐border information flows, the globalization of mass communication and the role of social media. All of these (and many other) channels may have contributed to reshaping national values/identities as well as trust and affinity towards foreign cultures (see Tabellini, 2008; Giuliano & Nunn, 2020; Bazzi et al., 2020).⁴ Of course, those changes in bilateral cultural proximity may or may not be reciprocated. The symmetric nature of cultural proximity is obviously very difficult to support, especially in the context of international migration. Symmetry in cultural proximity would imply, for instance, that the cultural barriers faced by Moroccans willing to move to France would be the same as those experienced by French migrants going to Morocco. As pointed out by Shenkar (2001), there are no studies showing symmetry in bilateral cultural proximity, nor is there any reason to assume symmetry. The use of the standard proxies of cultural proximity clearly fails to account for these dimensions of cultural relationships and their consequences for the migration decision. This calls for further investigation into the role of culture as a determinant of migration patterns.⁵

In this paper we propose an enriched notion of cultural proximity (henceforth VACP—Time-Varying and Asymmetric Cultural Proximity), which accounts for changes in cultural relationships that may or may not depend on the historical or pre-existing cultural ties. In this conceptual framework, the transfer of norms, practices, identities and social capital through social remittances, as well as exposure to foreign values and behavior, may change how attractive would-be migrants find foreign cultures. This is regardless of pre-existing bilateral cultural ties (see for instance Levitt, 1998). These “shocks” to bilateral cultural proximity affect the migration choice as, for any given country of origin, they alter the distribution of relative cultural affinity towards potential destinations.

To date, this is the first analysis exploring the relationship between cultural proximity and migration, which fully accounts for the time varying and asymmetric nature of VACP. In line with Disdier et al. (2010) and Fiorini et al. (2021), we employ cultural exports as a proxy for revealed cultural preferences. In other words, we assume that the value of bilateral exports of cultural goods reflects affinity towards the destination’s (exporter’s) culture for the citizens in the country of origin (importer). Our proxy for cultural proximity enters a gravity model of international migration which allows for cultural affinity to vary over time. Theoretically, we postulate that positive variations in cultural proximity, which are reflected by larger volumes of exports of cultural goods, lead to a rise in emigration from importing countries by lowering asymmetric bilateral moving costs between origin and destination.

On the empirical level, relaxing the assumption of stability of cultural proximity implies that migration could in principle affect the evolution of cultural affinity over time. Furthermore, current levels of cultural proximity are likely to be strongly related to historical cultural ties (see Disdier et al., 2010) and previous migration flows, introducing some endogeneity concerns. Our identification strategy addresses potential issues deriving from multiple sources of endogeneity by first instrumenting exports of cultural goods with average bilateral tariffs in the manufacturing sector and the imputed tariff revenues, which are plausibly exogenous with respect to migration. Second, to the best of our knowledge our identification strategy is among the first to utilize a comprehensive set of fixed effects—namely origin*time, destination*time and origin*destination Fes—within a gravity model applied to international migration. This enables us to estimate the impact of time varying cultural proximity on emigration over and beyond the effect of pre-existing cultural and historical ties. Lastly, in our gravity specification we separately identify the impact of existing diasporas as they simultaneously affect the decision to migrate both through cultural proximity, via the effect of cultural remittances. They also affect decisions to migrate by lowering migration costs through network effects and visa costs or by increasing the probability of non-economic migration through family reunification programs (see Beine et al., 2011).

The results suggest a positive impact on the time variance of cultural proximity on migration choice. In other words, positive changes in cultural proximity foster migration. This finding is robust across different econometric techniques and alternative classifications of cultural products. We also show that a shock in terms of changing cultural proximity has a much stronger effect on culturally distant country pairs and when the long-lasting effect of cultural goods in favoring cross-cultural convergence is accounted for. Taken together, these findings suggest a non-linear effect of cultural proximity on migration over pre-existing cultural ties and the potential (positive) role of trade integration in cultural products in promoting cultural alignment between origin and destination countries. Finally, the analysis not only supports the dynamic dimension of cultural proximity in the context of international migration, but also its asymmetric nature. Our findings reveal that only the time variation of migrant preferences for destination’s culture appears to affect emigration decisions.

The rest of the paper is organized as follows. Section 2 presents an overview of the related literature, while Sect. 3 introduces our enriched definition of cultural proximity and describes the theoretical framework. Section 4 outlines the econometric specification as well as the data utilized in the empirical analysis. Section 5 presents the main statistical results. Section 6 concludes.

Related literature

Our contribution adds to the extensive literature on the determinants of international migration, which uses gravity models as the main empirical workhorse to identify the effect of those origin, destination and dyadic factors affecting migration decisions. Gravity regressions first became very popular in analyzing international trade because they predict bilateral trade between two countries as a function of the respective economic sizes and distance between them. The theoretical foundations of gravity models of trade are widely explored in the literature (see Head and Mayer 2014). However, the interest in gravity models when applied to international migration “has only recently regained momentum because of an enhanced availability of migration data'' (Beine et al. 2015). Within this strand of literature, our empirical framework is similar to Ortega and Peri’s (2013) framework, which employs a comprehensive set of fixed effects and finds that international migration flows are highly sensitive to income per capita at destination and to bilateral migration policies. As stressed by Bertoli and Moraga (2013), the inclusion of an appropriate specification of origin and destination fixed effects into the gravity setup accounts for the so-called multilateral resistance to migration i.e. for the fact that the choice of a potential migrant to move to a given destination country does not only depend on the attractiveness of the destination country relative to the country of origin, ‘‘but also on how this relates to the opportunities to move to other destinations” (Bertoli & Moraga, 2013, p.79). Closer to the subject matter of this paper, Belot and Hatton (2012) show that cultural similarities and physical distance are more relevant drivers of educational selectivity in immigration than wage incentives or bilateral migration policies. A common feature of this strand of literature is that the effect of cultural distance on migration is mostly captured by dummies for common language (official or spoken) and former colonial ties (see Beine et al., 2015). A notable exception is Belot and Ederveen (2012) who capture different aspects of cultural similarities through the use of composite indicators for cultural proximity, along with more standard measures of cultural barriers such as religious and linguistic affinity. Similarly, Guiso et al. (2009) include, among the proxies of cultural similarities, measures of religious, linguistic, genetic and somatic distance. Among these proxies, which are however static and symmetric, linguistic distance has attracted particular attention. In particular, Adserà and Pytlikovà (2015) constructed elaborate indexes of linguistic distance and found that migration rates are higher between countries whose main official languages are closer. They also asserted that linguistic proximity matters less when local linguistic networks are larger.

Even though all of these measures aim to capture multiple dimensions of cultural similarities, an approach which is more in line with a comprehensive notion of “culture” (see Straubhaar, 2002), they implicitly assume that cultural proximity is constant over time and symmetric. In particular, the importance of the role of persistent/historical cultural traits on various political ramifications (Bazzi et al., 2020; Vertovec, 2011) and economic exchanges (Guiso et al., 2006) is well established in the literature. However, the role of the evolution of cultural proximity over time is often neglected and overlooked. To stress this argument even further, in reviewing the literature on gravity models for international migration, Beine et al. (2015) explicitly stated that cultural proximity is one of the most important “time invariant” dyadic components of bilateral migration costs. This definition of cultural proximity limits the capacity to capture all of the important dimensions of cultural affinity which have already been questioned in the international business, anthropological, sociological and economics literature (see Shenkar, 2001; Li et al. 2017; Felbermayr & Toubal, 2010; Giuliano & Nunn, 2020; Fiorini et al. 2021; Tung Verbeke, 2010). In this regard, Felbermayr and Toubal (2010). In this regard, Felbermayr and Toubal (2010) used the voting results of the Eurovision Song Contest (ESC) as a proxy for cultural proximity and found a significant time variation in the ESC scores that were awarded. They also evidenced a sometimes low degree of reciprocity even between countries with seemingly similar cultural attributes. Disdier et al. (2010) were the first to utilize trade in cultural products as a proxy for countries’ cultural proximity. They found that countries with similar cultural tastes have more intense trade relationships. Fiorini et al. (2021) combine these two contributions by applying cultural trade to study its impact on FDI. Our analysis employs a similar conceptual framework to study the impact of the time variation of cultural affinity on international migration by using bilateral exports in cultural goods from OECD to sending countries (migrant origins) as a proxy cultural proximity.

This paper also sheds some light on the role of trade in cultural goods in favoring migration through a progressive cultural alignment between origin and destination countries. We postulate that bilateral cultural exports raise reverse emigration by transferring cultural values, practices, habits, and norms to importing countries, making trading-pairs culturally closer and therefore lowering moving costs of would be emigrants. To date, the literature has mostly focused on the transfer of behavioral and cultural norms as a “migration externality”—i.e. the effect of diaspora networks on cultural integration through the role of social and cultural remittances (Levitt, 1998). For instance, Spilimbergo (2009) found that foreign trained individuals promote democracy in their home countries, but only if foreign education was acquired in a democratic destination. Focusing on Moldova, a former Soviet Republic, Barsbai et al. (2017) show that emigration to democratic countries decreases the share of votes for communist parties in home districts. Rapoport et al. (2020) found that migrants act as vectors of cultural diffusion and bring about cultural convergence through the dissemination of cultural values and norms from host to home countries (i.e., cultural remittances). In this paper we take a different approach and look at the effect on migration of cultural change induced by trade integration of cultural products. This links our paper to the work of Maystre et al. (2014), who show that trade integration leads to cultural convergence, especially for trade in differentiated goods with a higher cultural content. Along similar lines, we argue that cultural exports act as a sort of cultural remittances—as they transfer knowledge and familiarity about cultural values, practices and norms of migrant destinations to importing countries. In line with our hypothesis, the results presented in one of the extensions of the baseline model (Table 7) suggest a long-lasting positive effect of cumulative cultural exports on reverse emigration, which complements previous research on trade-based cultural change and its impact on emigration decisions (e.g. Maystre et al., 2014; Campaniello, 2014).

Cultural proximity and trade in cultural goods

Numerous empirical studies employed proxies of cultural affinity which overlook its time varying and asymmetric dimensions. For instance, the pioneering work of Belot and Ederveen (2012) employed several refined measures of cultural distance. This included a composite index based on the four Hofstede’s cultural dimensions, along with measures of linguistic and religious proximity, to estimate the impact of cultural barriers on international migration. The drawbacks associated with the definition of cultural distance have been brought to the fore in economics and even more prominently in other disciplines such as psychology, anthropology and the international business literature.⁶ In particular, two limitations of this approach stand out, namely the assumption of stability and symmetry, relabelled by Shenkar (2001) as, respectively, the “illusion of stability” and the “illusion of symmetry”.

The roots of the dynamic or processual view of cultural relationships that allows for values and practices to be continuously reshaped by social (and economic) interactions, can be traced back to the so-called “Intergroup Contact Theory” (Allport, 1954). The rationale behind the contact hypothesis is that more frequent and intensive contacts between different groups lead to a more tolerant society and improve social relations (also) through a reciprocal transfer of knowledge and information.⁷ The contact hypothesis implicitly relates to a pattern of cultural relationships that evolve over time, although not necessarily in a symmetric fashion. Closer to our purposes, the role played by diaspora networks through social and cultural remittances, for instance, is an example of how higher levels of cross-cultural interactions between host and origin communities can positively affect the perception and the affinity towards foreign cultures over time, through the so-called information or “acculturation” channel (Berry, 1980; Levitt, 1998). The static notion of cultural distance is also challenged by numerous anthropologists and sociologists, who propose definitions of culture that emphasize its dynamic, changing and transformative nature (Vertovec, 2011). While in these fields of research a micro-level approach is regarded as better-suited to study the evolution of cultural relationships, for the purpose of this study a notion of bilateral time-varying cultural proximity can still be introduced at the cross-country level—conditional on the assumption that national cultures share (at least to a certain extent) common cultural traits (see Guiso et al., 2009).

Hereafter, we briefly discuss each limitation in the context of international migration. We provide an alternative and purposefully broader definition of cultural proximity which allows for both time variation and asymmetry in cross country cultural relationships.

Stability The four dimensions identified by Hofstede (2001) measure how far apart two cultures are, as well as other standard proxies of bilateral cultural proximity such as religious and linguistic proximity. These are measured, though, at a single point in time and they are assumed to be constant. However, cultural proximity evolves over time. The cultural affinity towards a specific destination measured at the moment of the decision to migrate may have changed by the time cultures are measured. In other words, the distribution of the destination culture’s attractiveness across foreign countries changes over time; it is a function of several factors, including ideas and practices transferred to countries of origin. This variation in cultural proximity affects the migration choice as—for any country of origin—it alters the relative cultural affinity towards potential destinations.

Symmetry The construct of cultural distance obviously requires symmetry: the distance from A to B must be identical to the one from B to A. But this assumption has found no support in the literature. As highlighted by Shenkar (2001), Tung and Verbeke (2010) and Li et al. (2017) cultural relationships, which are relevant in the context of international economic exchanges, are far from being symmetric. Furthermore, Fiorini et al. (2021) and Felbermayr and Toubal (2010) found evidence of cultural asymmetry between country pairs. Their evidence points to an important role in the asymmetric dimension of cultural affinity in determining cross-country socio-economic interactions and calls for a broader notion of cultural proximity capable of reflecting asymmetric affinity between two countries. While the primary focus of this paper is on the role of a specific direction of time-varying cultural proximity—i.e. the affinity towards the destination’s culture for citizens in the country of origin—in one of the extensions we also analyze and quantify the implications of the asymmetric nature of cultural proximity in the context of international migration.

Our definitionIn line with these approaches, we depart from the construct of cultural distance and propose a workable definition of cultural proximity that relaxes both these assumptions. We assume that members of the same national culture share common cultural traits and have a fairly homogeneous view on the attractiveness of other cultures (Li et al. 2017, Brewer & Brown 1998).

We define cultural proximity as:

$$VACP_{in,t} = f\left( {G_{in} ,A_{in,t} } \right)$$

(1)

where $f$ is an increasing function of cultural proximity, ${G}_{in}$ denotes the time invariant component of cultural proximity. It stands for pre-existing or historical cultural ties, whose proxies, such as past colonial relationships, linguistic, religious and genetic distance, have been extensively used in the literature to capture the impact of cultural barriers on migration (see Beine et al. 2015). Contrary to the model proposed by Fiorini et al. (2021), $G_{in}$ may or may not be symmetric. The key assumption here is that the parameter $G_{in}$ is time invariant, so that $G_{in} = G_{in,t}$. $A_{in,t}$ denotes the attractiveness of n’s culture for the population in country i. $A_{in,t}$ is time varying and asymmetric, i.e. the identity $A_{in} = A_{ni}$ may not be verified at any time t. The evolution of cultural proximity over time for any country pair depends on the $A_{in,t}$ term, which may or may not be related to pre-existing cultural ties, i.e. $A_{in,t}$ could potentially not depend on $G_{in}$. Indeed, individuals in country i can, in principle, attribute desirable characteristics to the culture of country n independent of any actual similarity between the two cultures.⁸

Building on Disdier et al. (2010) we argue that bilateral trade in cultural goods can be used as a valid proxy for cultural proximity. More precisely, we employ the volume of cultural exports as a measure of “revealed cultural preference”—i..e. an increase in the demand of cultural goods from a given country of (migrant) destination reflects higher preference for the culture of that country. We postulate that positive variations in cultural proximity lead to a rise in emigration from importing countries by lowering asymmetric bilateral moving costs between origin and destination. The positive effect of VACP on migration operates mostly through the reduction of psychological costs and all the moving costs associated with settling in and adjusting to a new cultural environment, which—according to Kennan and Walker (2011)—play a crucial role in emigration decisions.⁹

Figure 1 plots the relationship between bilateral emigration rates from the importer country and the share of cultural exports from the country of destination. Our conjecture is that the correlation between migration and cultural exports is stronger because cultural preferences plausibly affect the utility derived from the purchase of cultural goods relatively more than their non-cultural counterparts. In other words, cultural preferences are better reflected in the purchase of cultural goods. In line with our hypothesis, Fig. 1 shows a positive relationship, which we attribute to the better capacity of cultural products to capture cross country cultural attractiveness.

Fig. 1

Log–log relationship between the bilateral share of cultural exports at t − 1 from the country of destination and the migration rate from the country of origin at time t. The right panel reports yearly correlations while the left panel shows the correspondent fitted line for the pooled sample. The share is constructed as bilateral cultural exports over aggregate bilateral exports. We exclude from the sample potential outlier observations as the information that belongs to the first and the last 3 percentiles of the distributions are dropped

The time-varying and asymmetric nature of trade flows allows us to these dimensions of bilateral cultural proximity. Specifically, given the purpose of this paper we are interested in cultural exports from destination n to origin i which proxies for ${A}_{in,t},$ i.e. the attractiveness of n’s culture for individuals in i at time t.¹⁰ The underlying theoretical mechanism is in line with the approach of Guiso et al. (2009). This is because changes in cultural proximity over time are reflected through variation in the demand for cultural goods produced in foreign countries. The recent case of the Korean Wave (Hallyu) is a good example of how the appreciation and penetration of the Korean culture in foreign countries is reflected in the rise of exports of cultural products, such as TV soap operas and pop music records (K-pop). Interestingly, the rise in popularity of the Korean culture around the world is not rooted in ancestral cultural traits but instead can be associated with a cultural shock (Fiorini et al. 2021). Scholars have started to identify a connection between the Korean Wave and a rise in economic exchanges. Box 1 in the "Appendix" highlights the case of the Korean Wave and illustrates its relationships with emigration decisions from importing countries.

Model and econometric specification

Econometric specification

The econometric model can be expressed as:of bilateral cultural flows

$$ln\left( {M_{in,t} } \right) = S_{i,t} + S_{n,t} + S_{in} + ln\left( {Xcult_{ni,t} } \right){ } + ln\left( {Q_{in,t} } \right) + u_{{{\text{in}},{\text{t}}}}$$

(2)

where ${Xcult}_{ni,t}$ is the bilateral exports of cultural goods from the destination country n to the country of origin i at time t. ${Q}_{in,t}$ is the bilateral stocks of emigrants born in i and resident in n at time t; it is included to proxy for migrant networks (diaspora) that reduce the cost of moving to destination country n. Networks may also be an indicator of cultural proximity because larger immigrant communities are likely to be associated with common cultural characteristics between hosting and origin countries. They can also be agents of change by transferring information. Our econometric specification is rich enough to be consistent with more general distributional assumptions of the error term ${u}_{\mathrm{in},\mathrm{t}}$ (see Beine et al., 2015). The inclusion of ${\mathrm{S}}_{\mathrm{i},\mathrm{t}}$ and ${\mathrm{S}}_{\mathrm{n},\mathrm{t}}$—which are respectively, origin-year and destination-year fixed effects—completely absorbs the impact of monadic (country-specific) determinants of international migration, such as conflicts as well as demographic, socio-political and environmental factors. In addition, these fixed effects allow us to account for the so-called multilateral resistance to migration. Put in other terms, it is possible to capture the impact of the influence that the attractiveness of possible other destinations exerts on the decision to migrate to a given destination.¹¹ In particular, the variable ${\mathrm{S}}_{\mathrm{n},\mathrm{t}}$ absorbs the average time-varying tightness of migration entry laws in every destination, which was found to have a significantly negative impact on immigration flows by Ortega and Peri (2013). In order to better isolate the time varying impact of cultural affinity on emigration we also include asymmetric origin–destination fixed effects ${S}_{in}$ which absorb all bilateral specific factors affecting migration. For the purpose of this paper the inclusion of ${S}_{in}$ is important for two main reasons:

• S_in allows us to identify the effect of cultural proximity over and above symmetric and pre-determined bilateral cultural ties. The current level of cultural proximity between country-pairs is likely to be related to the “historical” component of their cultural relationship (see "Appendix A1"). This may not be entirely captured by the time invariant proxies of ${G}_{in}$ commonly used in the literature to control for moving costs, such as colonial ties or linguistic, religious and geographical distance. This implies that failing to account for initial conditions may lead to biased estimates due to omitted variable bias (see Baier & Bergstrand, 2007). ${S}_{in}$ completely absorbs the cross-sectional variability in our sample. Hence, we are able to focus on the impact of the time contingent cultural factors alone.

• The inclusion of dyadic fixed effects restores the cross-sectional independence of the error terms (see Bertoli & Moraga, 2015). Indeed, if we define $b(i)$ as a nest of countries $i$ characterized by similar levels of cultural proximity with $n$, a bilateral shock between $n$ and $i$ may introduce a correlation in the stochastic component of Eq. (6). For instance, the impact of a more restrictive visa policy towards Moroccans in the UK will affect the relative attractiveness of other potential destinations which we realistically assume to be highly dependent upon cultural proximity between Morocco and third countries (i.e. on whether or not they belong to the same nest $b(i)$). In other words, if the unobserved components that create interdependencies across cross-sections within nests are correlated with the included regressors, the OLS estimator will be biased and inconsistent. Bertoli and Moraga (2015) restored the cross-sectional independence of the error terms through the inclusion of origin-nest dummies. Similarly, this paper proposes a richer analysis in which we generate a nest for each country-pair through ${S}_{in}$, alleviating potential estimation problems deriving from an incorrect specification.

Endogeneity concerns

An issue arising when estimating Eq. (2) is the potential endogeneity of trade in cultural goods. The main concern is whether this covariate is correlated with an unobserved component. In addition, since migration and trade are likely to be closely connected, the correlation between the two variables might be due to, other than the omitted variables we do not control for, reverse causality. This means that migrants may promote trade with their country of origin as well as cultural convergence (see for instance Gould, 1994; Rapoport et al., 2020).¹²

Our analysis aims to address the endogeneity issue in four ways:

• We include a comprehensive set of fixed effects to control for unobserved dyadic time-invariant factors and unobserved time-varying country-specific factors that drive both cultural proximity and migration flows. We are able to properly identify the relationship between cultural exports and emigration through the inclusion of country-pair fixed effects. To the best of our knowledge, this is among the first contributions to apply a gravity model to international migration which combines a full set of destination*year, origin*year and destination*origin FEs in the spirit of Baier and Bergstrand (2007) and Disdier et al. (2010). Finally, to further alleviate the problems associated with omitted variable bias we include bilateral imports as well as non-cultural exports in the specification (see Table 8), which partially control for time varying bilateral contacts between destination and origin.

• In line with Aleksynska and Peri (2014), we use the fact that the value of bilateral trade labeled as “cultural” according to UNCTAD classification, $Xcult_{{{\text{ni}},{\text{t}}}} ,$ is equal to the aggregate bilateral trade $X_{{{\text{ni}},{\text{t}}}}$ multiplied by the correspondent share of bilateral cultural flows $\alpha_{{{\text{ni}},{\text{t}}}}$. Specifically, $Xcult_{{{\text{ni}},{\text{t}}}}$ = $\alpha_{{{\text{ni}},{\text{t}}}}$*$X_{{{\text{ni}},{\text{t}}}}$. Hence, by taking logs and using log properties, we can separate the effect into two terms: $ln\left( {Xcult_{{{\text{ni}},{\text{t}}}} } \right) = ln\left( {X_{{{\text{ni}},{\text{t}}}} } \right) + ln\left( {\alpha_{{{\text{ni}},{\text{t}}}} } \right).{ }$ The advantage of this type of specification is that it builds on previous studies examining the trade-migration nexus, which normally included the log of aggregate trade as a dependent variable or a dyadic control in a gravity setup, depending on the direction of causation.¹³ Second, in our pooled OLS setting, aggregate bilateral trade absorbs common factors that affect aggregate trade and migration. This allows us to isolate and disentangle the extra impact of cultural products on migration flows within the same specification.

• The variable of interest—namely exports of cultural goods—is predetermined with respect to emigration flows, which is likely to (at least) attenuate the issue of reverse causality. The same “lagged approach” applies to other controls such as the impact of immigrant stocks $\left({Q}_{in,t}\right)$, in line with the analysis of Beine and Parsons (2015).¹⁴

• We propose an IV strategy where we instrument the flows of exports in cultural goods with the average bilateral tariffs in the manufacturing sector (source WITS, World Bank), which are applied by the importer and the corresponding imputed tariff revenues. As far as we know, we are the first to implement an IV strategy to study the relationship between cultural proximity and migration flows.¹⁵ Similarly to Campaniello (2014), the identification strategy in this paper hinges upon the assumption that bilateral tariffs do not depend on migratory flows. In other words, we postulate that governments set the level of tariffs to affect only trade flows, but not migration inflows. The first stage statistics along with Fig. 2 provide some empirical support to this statement, as the under identification, the reduced form and weak identification tests conducted and presented in Table 4 all point to the strength and the validity of the instruments. Hence, we plausibly assume the relationship between tariffs and migration as indirect i.e. running from the tariff-related instruments through the endogenous variable.

Fig. 2

Correlation between average bilateral tariffs at time t-1 in the manufacturing (in %) and a log of cultural exports at time t-1 (bottom panel) and b log of bilateral migration outflows at time t (upper panel). The right panels report yearly correlations while the left panels show the correspondent fitted line for the pooled sample. We exclude from the sample potential outlier observations as the information that belongs to the first and the last 3 percentiles of the distributions are dropped. The correlations refer to the smaller IV sample which includes 148 countries of origin and 22 countries of destination for the period 2004–2013

Data

The analysis uses data on 30 OECD countries of destination and on 185 countries of origin in the period 2004–2013.¹⁶ The sample composition is similar to the work of Adserà and Pytlikovà (2015). It is more comprehensive than other contributions that focus on the impact of cultural proximity on migration decisions, such as Belot and Ederveen (2012). The complete list, along with a short description of the variables employed in the statistical analysis, is presented in "Appendix A2", while the summary statistics of the main variables (including instruments) are outlined in Table 1.

Table 1 Summary statistics

Our main variable of interest is trade in cultural goods. Trade data are from the BACI dataset of CEPII, which provides the bilateral values of exports in the HS 6-digit product disaggregation, for more than 200 countries since 1995. A crucial issue for our analysis concerns the definition of “cultural goods”. In line with the definitions provided by UNESCO and UNCTAD we define cultural products as those goods “conveying ideas, symbols and ways of life to those who consume them (some of which may be subject to copyrights), and whose production requires some input of human creativity” (UNESCO 2009; UNCTAD, 2010). At the empirical level, we use the classification of cultural/creative products proposed by UNCTAD. "Appendix A3" provides the reasons why we prefer this classification as the main workhorse for our empirical analysis, while in Table 13 we list all of the domains and product codes labelled as cultural/creative according to both UNCTAD and UNESCO classification.

In the IV analysis we instrument exports of cultural products with the average bilateral tariffs applied in the manufacturing sector and the imputed tariff revenues from cultural trade. The average of bilateral tariffs is obtained as the simple mean across EORA manufacturing sectors.¹⁷ As for the rest of the variables employed in the statistical analysis, the migration flows and migration stocks are from the OECD’s International Migration database.¹⁸ Since we are interested in the determinants of migration decisions we use the inflows of foreign population by nationality in a given year as the dependent variable. This definition implies that we are including “all foreign-born (or in some cases foreign nationals) who come to the country to reside there and not for temporary tourism, study, or business reasons” (Ortega Peri, 2013). We include the stocks of bilateral immigrants who are resident in the countries of destination among the covariates, since they capture the role of networks in shaping international migration flows (see Beine et al., 2015). Standard Proxies for migration costs, such as weighted distance, common language, former colonial relationships, common legal origin, are from CEPII, while GDP per capita in PPP Constant US dollars are from the World Bank. More refined measures of pre-determined cultural proximity such as linguistic and genetic distance are from Adserà and Pytlikovà (2015) and Melitz and Toubal (2014).

Results

Benchmark estimates

The estimates of Eq. (2) are presented in Table 2. We progressively allow for lower degrees of variability across specifications in our identification data by gradually augmenting the number of fixed effects. Column (1) includes a reduced set of origin-year and destination dummies which capture time-varying factors at origin and time-invariant factors at destination, including unobserved heterogeneity in cultural traits between migrants and non-migrants. This specification is very close to the predictions of the model proposed by Ortega and Peri (2013). Our parameter of interest, the coefficient of $\mathrm{ln}({Xcult}_{\mathrm{ni},\mathrm{t}-1})$, suggests a significantly positive relationship between proximity of country i towards country n’s culture and bilateral emigration from origin i to destination n. All of the gravity controls are significant and have the expected sign. Income per capita at destination is confirmed as an important driver of migration flows, while the network effect is positive and its magnitude is in line with previous studies (see Beine et al. 2011; Beine & Parsons, 2015; Bertoli & Fernandez-Huertas Moraga, 2015). This result corroborates with the consensus in the literature on diasporas as the most important dyadic determinants of migration flows. Controlling for heterogeneity at destination-year level leaves our results substantially unaffected. The inclusion of destination-year fixed effects in Column (2), meanwhile, does not alter the coefficients of any of the dyadic explanatory variables.

Table 2 Benchmark results: impact of cultural exports on the emigration rate

These two specifications, however, do not tell us whether the effect of cultural proximity on the migration choice is only driven by historical and pre-existing cultural similarities. In other words, we cannot detect whether the evolution of cultural relationships over time plays a role in affecting migration decisions, as the time invariant component of cultural proximity, ${G}_{in}$, may act as confounding factor for the impact of${A}_{in,t}$. To address this issue, we include dyadic fixed effects ${\mathrm{S}}_{\mathrm{in}}$ which control for all time invariant bilateral factors, such as geographic barriers and pre-existing cultural ties. The results reported in Column (3) suggest that the time-varying determinants of migration remain significant despite the loss of identification power due to the large number of fixed effects introduced. In particular, the network coefficient retains the positive sign, but it lowers considerably in terms of magnitude, with a semi-elasticity of approximately 0.09 and statistically significance at the 1% level. More importantly for our purposes, the evolution of bilateral cultural proximity over time emerges as a significant driver of international migration: a “positive shock” in cultural proximity represented by an increase in cultural exports by 10% leads to a 0.13% increase in the reverse bilateral migration rate after controlling for all the dyadic and time invariant factors affecting migration decisions. In other words, cultural attractiveness affects the migration choice over and above the pre-existing cultural similarities. The effect is not only statistically significant, but also quantitatively relevant. A simple back-of-the-envelope calculation shows that moving from the sample median to the 75th percentile of cultural exports, leads to around 165 more migrants per dyad. This roughly corresponds to an additional 29,343 international immigrants per destination, which is about 16% of the average number of migrants per destination in 2010. This sheds some light on the importance of accounting for the evolution of cross-country cultural relationships and their linkages with recent migration phenomena. For instance, the 41% increase in international migrants from 2000 to 2014 may at least partially be explained by a trend of cultural convergence associated with globalization.¹⁹ Our results are consistent with such an interpretation. The last two Columns of Table 2 enrich the gravity specification by, respectively, adding non-cultural exports (Column 4) as additional control and decomposing exports in cultural products into the share of cultural products and total bilateral exports (Column 5). The findings suggest that only the time-variation of exports is positively associated with emigration flows, with the share of cultural exports having an impact above and beyond the correspondent aggregate flows.

The results hold when estimating the gravity equation with PPML (Table 3), which provides consistent estimates in the presence of heteroscedasticity and performs well when the dependent variable has a relatively large share of zeros (Santos Silva & Tenreyro, 2006, 2011; Bertoli & Moraga 2015). In our sample the share of zeros is rather small, it represents only 6% of the observations.²⁰ Despite some discrepancies in terms of magnitude with respect to the OLS counterparts, the PPML coefficients shown in Table 3 generally have the expected sign. More importantly, in line with our hypothesis, the impact of bilateral exports on migration seems to be predominantly driven by flows of cultural products. To further test the validity of our results, we estimate the gravity model with alternative econometric techniques such as Gamma PML and EK Tobit (Columns 5–6) which accounts for the zero migration flows. Although we cannot compare the performance of these estimators with high dimensional fixed effects, we find it reassuring that the estimates are in line with the results presented in Table 2.²¹

Table 3 Robustness check: alternative estimators

The results presented in Tables 2 and 3 are consistent with different sets of fixed effects and across econometric techniques. However, the reported estimates may still be biased because of reverse causality. To further address the potential endogeneity of trade in cultural goods we instrument $ln\left({Xcult}_{ni,t}\right)$ with the average bilateral tariffs in the manufacturing sector applied by the importer and the correspondent imputed tariff revenues (Table 4). Our hypothesis is that governments set the level of tariffs to affect only trade flows, but not migration inflows i.e. we assume that both tariff-related instruments affect migration indirectly i.e. only through their direct effect on the endogenous variable. Figure 2 provides some empirical support to this conjecture, as the average bilateral tariffs in manufacturing appear to be very weakly correlated to average migration flows.

Table 4 Robustness check: 2SLS results

The sample size for this IV exercise is reduced due to the tariffs dataset which does not provide information on all the country pairs included in our OLS sample.²² Table 4 reports the IV results. As expected, both the average bilateral tariffs and the imputed tariff revenues have the expected sign and are strong predictor of exports of cultural products. The Kleibergen-Paap F statistic of the excluded instruments is way above the conventional level and indicates that the instruments are well identified. Then we use the Hansen J-statistic to test the exogeneity and we find a p-value equal to 0.46, which points to the validity of our set of IVs. The reduced form in Column (3) suggests a direct relationship between the instruments and the dependent variable. By combining the first stage with the reduced form results (Columns 2–3) we can cautiously conclude that the effect of both instruments on the dependent variable runs through the endogenous variable. Of course, bilateral tariffs in the manufacturing sector are also related with non-cultural trade flows. This relationship might weaken the validity of our set of instruments if non-cultural trade is related with emigration flows and not accounted for in our model (Eq. 2). To address this issue, we perform the IV analysis by including non-cultural exports as additional control in our specification (Columns 4–6). The statistics suggest that while the time-variation of non-cultural exports is correlated with cultural trade flows (Column 4), it does not significantly affect emigration from importing countries (Column 5). The latter finding is in line with the baseline estimates reported in Column 5 of Table 2. All in all, the IV results essentially confirm the positive relationship that emerges from our baseline estimates and add consistency to our predictions on the importance role of cultural changes in the emigration decision.²³

Further addressing the measurement error bias

Measurement error can bias the estimated impact of our parameters of interest. While the use of trade in cultural goods as proxy for CP has many advantages for the purpose of this analysis, there are potential concerns regarding its validity in reflecting national cultural contents.

For instance, American music labels might export records with non-American cultural content, so the imports of music from the US in some cases doesn’t necessarily affect the perceived attractiveness towards the US culture. By the same token, French exports of fashion products (included in the UNCTAD classification of "optional" cultural goods) may not only reflect French cultural content, but also a third country’s cultural content embedded in the fashion design that is performed before manufacturing takes place in France (see Fiorini et al. 2021). Further, custom data does not include digital transactions (i-tunes, Netflix) that accounted for a relevant share of transactions of several "core" cultural goods, such as DVDs, Music and Books. However, digital transactions have increased dramatically over the last 5–6 years, a period that falls outside our sample’s time coverage, so the latter source of measurement error is unlikely to largely influence our results.²⁴

To address the issues associated with measurement error in Table 5 we first compare the benchmark findings reported in Table 2 (Column 1) with the correspondent estimates obtained with the “core” UNESCO classification of cultural products (Column 4–5). The products identified by UNESCO as cultural goods are arguably characterized by a larger cultural content compared to the classification proposed by UNCTAD. They are therefore likely to better capture proximity in cultural tastes. However, as noted in "Appendix A3", UNESCO’s classification implies the use of a more limited time span and is less representative of the cultural products traded by the South. Given the shorter time coverage we are not including our full set of FEs since the more limited information in the UNESCO sample would create problems in terms of identification power. Hence, we compare the two classifications only with country-year fixed effects. The results indicate that using a different classification does not alter our benchmark estimates and our main conclusions remain unchanged. Lastly, in Column (2–3) we propose trade in newspapers and other printed matter as a more refined/accurate alternative measure of cultural proximity (see "Appendix A3" for more details on these product categories). The idea behind this is that newspapers are less subject to the global value chain bias described above, as their production is not dislocated to foreign countries. This therefore minimizes the potential concerns regarding the measurement error introduced by the gross nature of cultural trade. The results point to a positive relationship between cultural changes and emigration, which corroborate our baseline estimates.

Table 5 Robustness check: UNCTAD versus UNESCO classification

Extensions

This section proposes two extensions to the analysis conducted so far. We test whether the role of the time varying component of cultural proximity changes (a) at different levels of pre-existing cultural similarities and (b) when we account for the long-lasting effect of cultural goods in favoring cross-cultural convergence.

Table 6 explores the variation of the role of cultural proximity on emigration for different levels of pre-determined cultural affinity and stages of economic development.

Table 6 Extension: countries’ development status and initial cultural similarities

We first divide our sample according to the degree of cross-country cultural affinity based on linguistic and genetic distance (Columns 1–4) as well as the average volume of cultural exports (Columns 5–6). In order to preserve enough identification power and to attenuate the selection bias we split the sample into, respectively, almost identical subgroups using the median of fst_distance_dominant from Adsera and Pytlikova (2015), lp2 from Melitz and Toubal (2014) and the average value of cultural exports over our period of interest, respectively.²⁵ Taken together, the results suggest that time contingent shocks to cultural proximity only play a role when historical cultural similarities between country pairs are relatively weak. This finding suggests a non-linear effect for cultural proximity on migration over pre-existing cultural ties and a potential role for trade in cultural products in promoting cultural convergence.²⁶ In particular, the evidence is consistent with a relationship of substitutability between the time-contingent, asymmetric and time-invariant, symmetric dimensions of cultural proximity in triggering migration, with the former operating as a bridgehead between otherwise culturally distant countries. A plausible explanation is that the cultural content embodied in these types of products enhances bilateral cultural affinity through what Tabellini (2008) defines as the horizontal transmission of values. The consumption and diffusion of cultural goods in countries of origin can contribute to transferring exporter’s cultural traits, making the culture at destination better known, more attractive and more widely accepted.

In Columns (7–8) we test whether positive shocks in CP influence migration between country-pairs at different stages of development. To do so, we split the sample according to what is typically classified as North–North vs South-North migration and define as North countries all the member states of the OECD included in our sample. Interestingly, the estimates suggest that the effect of the time-variation of cultural proximity comes from South-North migration. In other words, a positive shock of VACP—other factors held constant—raises emigration towards countries characterized by larger income differentials. While this finding corroborates the results reported in Columns (1–6)—as we expect that the cultural distance between OECD countries and non-OECD countries may be larger than between OECD countries—it also suggests that the role of cultural proximity in reducing moving costs appears to be particularly important in developing countries, where budgetary and credit constraints are more binding. Finally, when looking at the impact of diaspora across sub-samples, it appears to be stronger for North–North migration. This result is in line with the literature on the role of networks in micro-founded gravity models, as the elasticity of the stock of emigrants generally increases when focusing on emigration towards OECD destinations (see Beine et al. 2015). In addition, we do not consider this evidence at odds with our hypothesis—given that a larger diaspora coefficient might be explained by the skill composition of networks (Felbermayr & Jung, 2009), for which we do not have data that fully cover our sample’s time-span.²⁷

While in this study we are employing cultural exports mainly as a proxy for “revealed cultural preferences”, we are not ruling out the cultural transmission channel of cultural trade (Maystre et al. 2014), i.e. that cultural content embodied in cultural goods can transmit and diffuse information on values, beliefs, habits and cultural traits of migrant destinations in importing countries. This process would in turn lead to a rise in emigration from importing countries through a progressive cultural alignment between origin and destination countries. In Table 7 we explore more closely this potential long-lasting role of trade in cultural goods in favoring cross-country cultural convergence and its indirect impact on the decision to migrate. More specifically, we test whether the intensity of long-lasting bilateral cultural relationships have a stronger effect on migration. We are well aware that the transmission of values which shapes the utility of would-be migrants takes time (see Cavalli Sforza, 2001).²⁸ For instance, the effect of traded movies on cross-country cultural pervasiveness is neither instantaneous or brief; rather, movies can be repeatedly watched and broadcast once purchased. Hence, our empirical strategy accounts for the recent history of bilateral trade relationships between $n$ and $i$ by simply considering the impact of the cumulative exports of cultural products from destination $n$, so that:

$$CumXcult_{ni,t} = \mathop \sum \limits_{t - 1}^{t - s} Xcult_{ni,t}$$

(3)

Table 7 Extension: impact of ‘"Cumulative’’ cultural exports on the emigration rate

This strategy allows at the same time to attenuate the distortion due to business cycle factors and measurement error associated with trade data. We initially set $s\hspace{0.17em}$ =  5 while the third column reports the correspondent estimates with $s\hspace{0.17em}$ =  9. Interestingly, as $s$ goes up the impact of cultural exports tends to increase. In other words, when we add past bilateral cultural exported goods to ${Xcult}_{ni,t-1}$ the impact of our variable of interest on the decision to migrate at time t gets larger and larger. This finding is consistent with the hypothesis of a long-lasting effect of cultural products on bilateral cultural affinity.

Table 8 deals with the asymmetric dimension of time-contingent variations of cultural proximity in the context of international migration. To do so, we add imports of cultural goods to our gravity specification. We start with a parsimonious specification including only cultural imports in the model (Column 1), then progressively add other variables, namely imports of non-cultural goods (Column 2), exports of cultural (Column 3) and non-cultural goods (Column 4). Generally, the evidence indicates that the asymmetric component of VACP matters. Taken together, the results suggest that only the time-variation of cultural exports have a positive and significant effect on emigration from importing countries—i.e. the preference of the individuals in the importing country for the exporter’s culture appears to be the only direction of cultural proximity that influences emigration decisions from importing countries. In other words—according to our conceptual framework—while a rise of, say, cultural affinity for Mexicans towards US culture leads to higher emigration to the US, the time variation of US preferences for Mexico’s culture does not appear to affect Mexicans’ emigration decisions. Finally, when restricting the analysis to the within variation of country-pairs both directions of trade in non-cultural products do not influence emigration flows.

Table 8 Extension: asymmetric cultural proximity

Conclusions

Cultural barriers have been identified as one of the main drivers of international migration. They explain patterns of international migration which cannot be explained by merely looking at differences in terms of economic indicators. In other words, cultural factors help to address the so-called “immobility puzzle”, which we can define—paraphrasing Trefler (1995)—as “the case of missing migration”, i.e. very low migratory responses to large unemployment and wage differentials. In this context, proxies for cultural proximity such as linguistic and religious distance, along with more refined indicators, capturing (at least partially) the cultural orientation of countries, were found to have a positive impact on migration flows, after controlling for income differentials.

However, the common characteristics of these proxies—and, more generally, one of the implicit assumptions adopted in international migration gravity models—is that cultural proximity is time invariant and symmetric. These assumptions appear particularly strong and unrealistic when looking at the migratory patterns of the last two decades, given the growing exposure of citizens to foreign cultures through cross‐border information flows, the globalization of mass communications and the rise of social media. All these channels may have affected the degree of cultural affinity of citizens towards potential destinations, without this affinity being necessarily reciprocated.

In this paper we relax these assumptions and propose a broader definition of cultural proximity which explicitly accounts for the asymmetric evolution of cross-cultural relationships over time. In line with Disdier et al. (2010) and Fiorini et al. (2021), we use bilateral trade in cultural goods as a proxy for time-dependent and asymmetric cultural proximity. More precisely, we assume that the value of the bilateral exports of cultural goods reflects affinity towards the destination’s (exporter’s) culture for the citizens in the country of origin (importer). Our analysis contributes to the literature as the impact of the time-varying and asymmetric dimension of cultural proximity was too little studied in the context of international migration. The few existing studies on the impact of cultural barriers on migration choices were, meanwhile, predominantly focused on OECD countries as the point of origin of migrants, leaving out the whole spectrum of developing countries where the impact of cultural proximity on the decision to migrate might be particularly relevant.

Relying on a comprehensive set of high dimensional fixed effects and controlling for the size of diaspora, we find that the time variance of cultural proximity helps explaining international migration. More specifically, when accounting for pre-existing bilateral cultural ties, an increase of cultural affinity towards a potential destination enhances bilateral migration flows towards that destination. The results are robust across several robustness checks, including an IV strategy where exports of cultural goods are instrumented with tariff-related instrumental variables. Interestingly, when decomposing total exports between cultural and non-cultural flows, we find that the time variation of cultural exports seems to have a much more prominent role in explaining international migration patterns. This finding not only supports the validity of cultural exports as a proxy for VACP—whose impact on reverse emigration goes over and beyond the effect of aggregate trade flows—but also complements previous research on the trade-migration link (e.g. Campaniello, 2014) by distinguishing different types of trade flows. More specifically, when restricting the analysis to the within variation of country-pairs, trade in non-cultural products does not appear to influence emigration decisions from countries of origin. Furthermore, our findings suggest that the reverse channel of VACP—proxied by cultural imports—does not explain emigration flows. In other words, only the time variation of migrant preferences for destination’s culture appears to affect emigration decisions. This result supports our theoretical considerations and corroborates the findings of Fiorini et al. (2021) and Felbermayr and Toubal (2010) on the asymmetric nature of cultural proximity. Finally, the positive impact of cultural proximity is found to be even stronger for culturally diverse country pairs, and when the long-lasting effect of cultural goods in favoring cross-cultural convergence is taken into account. The latter finding sheds some light on the positive role of trade integration in cultural goods in favoring migration through a progressive cultural alignment between origin and destination countries.

This paper leaves interesting avenues for policy analyses and future research. First, more disaggregated analyses would constitute an important step towards arriving at more nuanced policy conclusions. For instance, exploring how the effect of cultural shocks on migration decisions vary with the level of importing countries’ cultural heterogeneity would help providing a more detailed picture of the link between cultural proximity and migration decisions. This of course implies relaxing the assumption of within-country cultural homogeneity. Further, empirical research is also needed to better understand how positive variations of cultural proximity influence emigration at different skill levels. One would expect a larger impact on the highly skilled, those who are more used to diversity and plausibly more open to external values. Second, our findings suggest a stronger positive impact from exports of cultural goods when historical cultural similarities between country pairs are relatively weak. The evidence is consistent with a relationship of substitutability between time-contingent and time-invariant dimensions of cultural proximity in triggering migration, with the former operating as a bridgehead between otherwise culturally distant countries. Further research should be conducted to understand the role of the trade in cultural goods in shaping cultural values and triggering cultural convergence. Lastly, policies and programmes aimed at promoting national cultures in foreign countries (e.g. Korea) may have unintended consequences on the expected costs associated with migrant integration. By lowering the psychological and moving costs associated with adjusting to a new cultural environment, these policies can potentially reduce future integration costs in destination countries as emigrants adapt more easily to the values and habits of the hosting communities (Berry, 1980).

Appendix: Cultural change and the migration choice

See Box1 and Fig. 3.

Box 1 The Korean wave “Hallyu”

Fig. 3

Korean cultural exports to selected Asian countries and immigration flows. Notes Correlation between Korean cultural exports to Philippines (top panel) and Malaysia (bottom panel) and the evolution of reverse bilateral emigration flows to Korea over time

.

Table 10 Korean cultural exports and immigration flows

A1 Link between cultural trade and time invariant proxies for cultural proximity

In this Section, similarly to Felbermayr and Toubal (2010) and Guiso et al. (2009), we show that imports of cultural goods reflect time-invariant components of cultural similarities. We also argue that their time and within country-pair variance are strongly associated to the broad definition of cultural proximity proposed by sociologists (see Felbermayr & Toubal, 2010).

Table

Table 11 Correlation coefficients between different measures of static cultural proximity

11 shows pairwise correlation coefficients between Imports of Cultural goods and standard pre-determined measures of cultural similarities. Among these measures we include the log of weighted distance, a dummy for a common colonial past (CEPII) and measures of linguistic and genetic distance from Adserà and Pytlikovà (2015). In addition we compute correlations with measures of the size of networks (OECD) along with the religious distance and the composite Hofstede Index from Belot and Ederveen (2012). In particular, this composite Index has been one of the main workhorses for the empirical test of the impact of cultural affinity on economic exchanges such as trade and FDI (see for instance Fiorini et al. 2021), but other than being at the same time pre-determined and symmetric, has the drawback of covering a fairly limited sample. The correlation coefficient between both cultural imports measures and the conventional proxies of cultural and geographical proximity all have the right sign and are statistically different from zero.

Table

Table 12 Explaining trade in cultural goods

12 reports the results of a simple regression which closely resembles to the gravity models proposed by Disdier et al. (2010) where trade in cultural goods is explained by time invariant measures of cultural similarities. Compared to Table 12 we also include imports of non-cultural goods among the explanatory variables. Interestingly, while all measures are strongly related to cultural imports in pairwise correlations, the regression analysis shows that religious and genetic distance as well as the size of the network does not matter for trade in cultural products. This might be due to the large impact of the Hofstede Index which plausibly captures much of the impact of other measures of cultural similarities. The results clearly indicate that Imports of cultural goods greatly depend on pre-determined components of cultural similarities.

A2 definition and sources

See Table

Table 13 Definition and sources of variables used in the empirical analysis

13.

A3 UNCTAD versus UNESCO classifications

A crucial issue for our analysis concerns the definition of what products can be considered as “cultural goods”. Two alternative classifications for cultural goods have been proposed by two different United Nations Organizations (UNESCO and UNCTAD), each of them based on slightly different criteria and different categories of goods to be included. As opposed to Disdier et al. (2010)—who classified cultural/creative goods according to the definition promoted by UNESCO—we rather adopt the scheme proposed by UNCTAD (2010). Our choice hinges on two main reasons: the first one—more substantial—concerns the countries’ selection; the second—more technical—is related to the different coding system adopted by the two classifications.

• Although the two frameworks apply different versions of HS codes, UNESCO and UNCTAD embrace some common underlying principles for capturing the exclusive creative/cultural goods. However, according to the data from UN COMTRADE database, the ‘’core’’ categories provided by UNESCO are dominated by developed economies, whereas the ‘’optional’’ domains more represented in the broader UNCTAD classification stand for the predominant share of cultural imports in the developing countries.²⁹ Since our analysis includes a very large number of developing and transition economies as countries of origin, we argue that UNCTAD classification is more adequate as it balances the cultural composition of trade flows by guaranteeing a relatively more comparable measure of cultural trade across countries at different development stages.

• The most recent and updated classification adopted by UNESCO is the 2007′s Harmonized Commodity Description and Coding System (HS 2007). The use of HS07 would imply a more limited time span as the first available year would be 2008, which would call for the adoption of conversion tables in order to extend the time coverage. One of the drawbacks of using conversion tables is that different levels of disaggregation across HS classifications may create distortions in the definition of “cultural products”. Conversely, UNCTAD (2010) adopts the HS02 coding system, which naturally allows for larger time series information. For the purpose of this paper the larger sample size is particularly important in terms of identification power given the important number of high dimensional fixed effects included in the specification, hence UNCTAD is preferred. In one of the extensions in Sect. 4.3 the cumulative exports are aggregated over 9 years, therefore conversion tables (HS92 to HS02) have been utilized.

Combining these two observations, the selection of cultural classification implies a trade-off between the cultural content embodied in these goods and the representation of developing countries as well as time coverage. To address this issue, we check the robustness of our results by employing the “core” UNESCO classification. The product codes in both classifications are outlined in the next Table

Table 14 Classifications of cultural products

14.