This paper identifies and estimates the causal effect of an intervention on repeatedly measured units that co-exist and interact with one another in a social network, when the dichotomous intervention is not randomly assigned and the network evolution may be driven by choices of social agents. We adopt the potential outcome framework and develop identification assumptions to define and identify three estimands, namely, the direct treatment effect, the spillover effect, and the general treatment effect.  Our framework incorporates  social network ties as part of the joint treatment and treats longitudinal networks as variables rather than constants. It also considers complicated causal paths generated by interdependent outcomes. We propose a model-based estimation strategy and use a factor analysis to correct for biases caused by latent homophily. By imputing potential outcomes based on simultaneous equations, we disentangle spillover effects from direct treatment effects and explicitly estimate first-order and higher-order causal effects. The proposed method is easy to implement and flexible to accommodate a wide variety of networks. 

How does China’s growing financial power reshape the structure of world politics? Does the expansion of China’s global influence come at the expense of the United States? A growing body of research in International Relations (IR) finds that China is increasingly using its economic power to gain leverage in international politics and engage countries to support its foreign policy. This paper addresses the question of whether “the rise of the Renminbi" – China’s entry into global monetary affairs through its BSA network – boosts its political influence and diminishes US political power on the global stage. To tackle the identification problem, we employ a Bayesian causal inference method devel-oped by Pang et al. (2021): bpCausal. Our analysis reveals intriguing results. First, we find that on average China’s BSAs have statistically a significant short-run causal effect of on governments’ expressed preferences over global issues. That is, BSAs bring the foreign policy ideal points of China and its partner states closer together in the year after a BSA is signed. Interestingly, while BSAs shift signatories closer to China’s ideal point in the UNGA, we find no evidence that BSAs pull countries away from the positions of the United States. On the contrary, weak evidence suggests that governments are moving closer to the United States in the median run, holding all other things constant, though the impact is much smaller compared to moving closer toChina.

The regime divide is one of the most studied cleavages in international politics, and the current discussion centers on whether the great power competition between the United States and China divides the world along regime lines. This paper focuses on the US-China competition in forming voting alignments in the United Nations General Assembly and disentangles the effects of regime type on actions, preferences, and strategic calculations of the rival powers and developing countries. We develop a formal model to theorize the competition and convert the game into a Bayesian statistical estimator. Empirical evidence suggests that the US-China competition increases the democracy/authoritarianism voting cleavage. States' regime-oriented voting or vote-buying choices, however, are not driven by their sincere preferences but by differential strategies shaped by regime type. These findings shed light on the nature of the US-China competition and its implications for the world order.

Disseminating data is a core mission of international organizations (IOs). IO lending and conditionality may incentivize governments to collect and disclose aggregate economic data. We explore the association between loans from the Bretton Woods Institutions (BWIs) and an index of economic transparency derived from the data-reporting practices of governments to the World Bank. Using a matching method for causal inference  complemented by a multilevel regression analysis, we find consistent evidence of a positive and statistically significant effect of BWI loans on the improvement of the level of economic transparency in developing countries. 

This paper proposes a Bayesian multilevel spatio-temporal model with a time-varying spatial autoregressive coefficient to estimate temporally heterogeneous network interdependence. To tackle the classic reflection problem, we use multiple factors to control for confounding caused by latent homophily and common exposures. We develop a Markov Chain Monte Carlo algorithm to estimate parameters and adopt Bayesian shrinkage to determine the number of factors. Tests on simulated and empirical data show that the proposed model improves identification of network interdependence and is robust to misspecifcation. Our method is applicable to various types of networks and provides a simpler and more flexible alternative to coevolution models.

This paper proposes a Bayesian alternative to the synthetic control method for comparative case studies with a single or multiple treated units. We adopt a Bayesian posterior predictive approach to Rubin’s causal model, which allows researchers to make inferences about both individual and average treatment effects on treated observations based on the empirical posterior distributions of their counterfactuals. The prediction model we develop is a dynamic multilevel model with a latent factor term to correct biases induced by unit-specific time trends. It also considers heterogeneous and dynamic relationships between covariates and the outcome, thus improving precision of the causal estimates. To reduce model dependency, we adopt a Bayesian shrinkage method for model searching and factor selection. Monte Carlo exercises demonstrate that our method produces more precise causal estimates than existing approaches and achieves correct frequentist coverage rates even when sample sizes are small and rich heterogeneities are present in data. We illustrate the method with two empirical examples from political economy. (For software to implement the method, please visit https://github.com/liulch/bpCausal)

Bilateral cross-currency SWAP agreements (BSAs) were rejuvenated during the GFC by the Federal Reserve of the United States to provide liquidity to selected governments in dire financial straits. After the crisis, BSAs rapidly and widely spread around the globe and become a major component of the global financial safety net. Interestingly, it is not the United States but China which occupies the center of the global BSA network. China has signed BSAs with a diverse pool of states since 2009, and has far more partners than any other countries. While existing studies mainly focus on the motivations driving China and its partner states to enter BSAs, this paper is intended to evaluate foreign policy consequences of China’s BSAs. Asymmetric interdependence and asymmetric information, as two key features of the economic relationship between China and its partner states in BSAs, are expected to make other countries be more supportive to China’s position in global affairs, leading to a convergence of their foreign policy preferences. This is the classic “Hirschman effect”. To empirically identify the “Hirschman effect”, we use measures of states’ foreign policy ideal points based on votes in the United States General Assembly. We apply a quantitative analysis to estimate the average effect and conduct a case study to trace and explain the development of the effect over time. In the large-N study, we draw data on 191 countries between 2009 and 2018 and specify a multilevel model with varying intercepts to control for unobserved heterogeneity in the dimension of time and space. Empirical evidence suggests that BSAs significantly drive the foreign policy preferences of China and other states to converge. Then we focus on Argentina as an in-depth case study. Different from conventional case studies, we conduct a “quantitative case study” and apply the Synthetic Control Method to estimate and quantify the causal effect of signing a BSA with China in 2009 on the distance between foreign policy ideal points of Argentina and China. The case study confirms the presence of the “Hirschman effect” suggested by theory and found in the regression analysis. It further reveals several suggestive but interesting points, including 1) activating SWAP lines may strengthen the effect, whereas the effect may be weakened by the provision of emergency liquidity assistance from the US or the IMF; 2) BSAs with China may impact on left-wing governments more strongly than right-wing governments; and 3) the Chinese government may strengthen the effect by changing the size of the committed SWAP line.

The US-China relationship is one of the most important bilateral relationships in today’s world politics. Conflict and cooperation between the two major powers affect regional and global stability and the dynamics of the international system. What are the patterns of the ebbs and flows of this relationship? And how can the US-China conflict and cooperation be explained by reciprocity, policy inertia, and influence of an important third party? Scholars have used machine-code event data and time-series tools to analyze the three factors in American-Soviet relations. This research extends the existing research and applies new data and methods to trace and explain the variations in US-China interactions. We rely on data from event data from the Global Data on Events Location and Tone (GDELT) and obtain a sample of directed actions of China, the United States, and Russia/U.S.S.R. towards one another between 1979 and 2017, totaling 3,957,479 daily records. We first apply multivariate change-point analysis and locate three structural breaks in the 38 years. For each of the four sub-periods, we specify a Vector Autoregressive Model and utilize the Impulse Response Functions to estimate the mutual effects of six time series of directed actions among China, the U.S., and Russia. Then we build signed and directed networks using country dyads as nodes and estimates as edges to summarize and interpret the interdependence among dyadic interactions in this triangle. The paper has three major empirical findings. First, the most important factor behind the US actions toward China is policy inertia that could reflect the effect of domestic politics on American foreign policy decision- making. But China’s action towards the US are equally determined by reciprocity, policy inertia, and Russia as the third party. Second, we find mutual reciprocity in the China-US interactions, but reciprocity is highly assymmetric---the responses of the US to China’s actions are much weaker than China’s responses to the US actions, which could explain why US-China cooperation is often difficult to reach. Thirdly, the China- US-Russia triangle is featured by a logic of “balance of power” --- Russia and China are more cooperative to each other when their relations with the US get more conflictual, and vice versa. The pattern shows that the cooperation between China and Russia is a tactic to increase their leverage to settle conflicts with the US. Furthermore, there is no stable reciprocity between China and Russia, which means that a China-Russia alliance is unlikely to become true despite that the foreign policy rhetoric and diplomatic gestures seem to suggest an emergence of such an alliance.

Existing scholarship on the rise of China and Chinese foreign policy has largely neglected to explain one puzzling phenomenon—although China’s network position in the global economic system has become more similar to that of developed countries, China continues to set its foreign policy from the perspective of a developing country. By analysing the relationship between the environmental possibilities and China’s intentional strategy, this article argues that the sharp contrast between China’s positions in the international political and economic systems reflects China’s ‘network strategy of embedded rise’. There are two mechanisms that go in opposite directions and jointly determine China’s foreign policy location in the international political spectrum. The first, called the ‘structural alienation effect’, is that wherein divergence in the centre-periphery positions in the global economic network increases the heterogeneity of two countries’ preferences, attitudes, and incentives, and in turn weakens alignment of their foreign policies. However, this structural effect is moderated by the second mechanism, namely, the ‘strategic affinity effect’. China’s strategy of assuming the role of a ‘broker’ to connect developing countries at the periphery of the international political and economic networks with developed countries at the centre achieves high ‘betweenness’ centrality, and so increases its social capital and influence in international politics. This article tests the proposed theory of China’s network strategy of embedded rise through hierarchical models that use global trade network centrality data and the United Nations General Assembly (UNGA) votes of 161 developing countries from 1994 to 2012. Our findings show that the globalized world both constrains and enables China’s great power status.

This essay joins the discussion on ‘‘The Future of Comparative Politics’’ from a perspective of methodology, and argues that the challenges concerned in Schmitter’s essay are not endemic to comparative politics but shared ones in other research fields including quantitative methods. Recent trends and developments in quantitative methods show that quantitative and qualitative methods are increas- ingly integrated to jointly handle challenges with broad and profound impacts on the social sciences as a whole. This essay presents a brief introduction of the recent three revolutions in quantitative methods. The ‘‘Bayesian Revolution’’, the ‘‘Credibility Revolution’’, and the ‘‘Big Data Revolution’’ have fundamentally changed quantitative methods. The paper further displays that the challenges arising from the three revolutions are essentially the same ones with those in comparative politics, such as modeling complex interdependence, dealing with fuzzy concepts and the messy real world, and so on. Finally, the essay uses a few examples of some new analytical tools developed by quantitative methodologists to illustrate that qualitative knowledge and quantitative techniques should be seamlessly mixed to be innovative and powerful methods. All this points to a common future of compar- ative politics and quantitative methods.

Following older debates in international relations literature concerning the relative importance of domestic versus systemic factors, newer debates emphasize interdependence among states and the complex interactions between systemic and domestic factors. As globalization and democratization advance, theories and empirical models of international politics have become more complicated. We present a systematic theoretical categorization of relationships between domestic and systemic variables. We use this categorization so that scholars can match their theory to the appropriate empirical model and assess the degree to which systemic factors affect their arguments. We also present two advances at the frontier of these empirical models. In one, we combine hierarchical models of moderating relationships with spatial models of interdependence among units within a system. In the other, we provide a model for analyzing spatial interdependence that varies over time. This enables us to examine how the level of interdependence among units has evolved. We illustrate our categorization and new models by revisiting the recent international political economy (IPE) debate over the relationship between trade policy and regime type in developing countries.

Multifactor error structures utilize factor analysis to deal with complex cross-sectional dependence in Time-Series Cross-Sectional data caused by cross-level interactions. The multifactor error structure specification is a generalization of the fixed-effects model. This article extends the existing multifactor error models from panel econometrics to multilevel modeling, from linear setups to generalized linear models with the probit and logistic links, and from assuming serial independence to modeling the error dynamics with an autoregressive process. I develop Markov Chain Monte Carlo algorithms mixed with a rejection sampling scheme to estimate the multilevel multifactor error structure model with a pth-order autoregressive process in linear, probit, and logistic specifications. I conduct several Monte Carlo studies to compare the performance of alternative specifications and approaches with varying degrees of data complication and different sample sizes. The Monte Carlo studies provide guidance on when and how to apply the proposed model. An empirical application sovereign default demonstrates how the proposed approach can accommodate a complex pattern of cross-sectional dependence and helps answer research questions related to units' sensitivity or vulnerability to systemic shocks.

Jurisprudential regime theory is a legal explanation of decision-making on the U.S. Supreme Court that asserts that a key precedent in an area of law fundamentally restructures the relationship between case characteristics and the outcomes of future cases. In this article, we offer a multivariate multiple change-point probit model that can be used to endogenously test for the existence of jurisprudential regimes. Unlike the previously employed methods, our model does so by estimating the locations of many possible changepoints along with structural parameters. We estimate the model using Markov chain Monte Carlo methods, and use Bayesian model comparison to determine the number of change-points. Our findings are consistent with jurisprudential regimes in the Establishment Clause and administrative law contexts. We find little support for hypothesized regimes in the areas of free speech and search-and-seizure. The Bayesian multivariate change-point model we propose has broad potential applications to studying structural breaks in either regular or irregular time-series data about political institutions or processes.

This paper proposes a Bayesian generalized linear multilevel model with a pth-order autoregressive error process to analyze unbalanced binary time-series cross-sectional (TSCS) data. The model specification is motivated by the generic TSCS data structure and is intended to handle the associated inefficiency and endogeneity problems. It accommodates heterogeneity across units and between time periods in the form of random intercepts and random-effect coefficients. At the same time, its pth-order autoregressive error process, employed either by itself or in concert with other dynamic methods, adequately corrects serial correlation and improves statistical inference and forecasting. With a stationarity restriction on the error process, the model can also be used as a residual-based cointegration test on discrete TSCS data. This is especially valuable because cointegration testing on discrete TSCS data is methodologically challenging and rarely conducted in practice. To handle the estimation difficulties, I developed an efficient Markov chain Monte Carlo (MCMC) algorithm by orthogonalizing the error term with the Cholesky decomposition and adding an auxiliary variable. The parameter expansion method, that is, partial group move–multigrid Monte Carlo updating (PGM-MGMC), is employed to further improve MCMC mixing and speed up convergence. The paper also provides a computational scheme to approximate the Bayes’s factor for the purposes of serial correlation diagnostics, lag order determination, and variable selection. Simulated and empirical examples are used to assess the model and techniques.