Financial integration is often perceived to lead to convergence of asset prices, as well as higher comovements across countries, with the idea that the dependence on world factors should increase as markets integrate. This dissertation focuses on analyzing how integration has changed over time in developed and, especially, emerging markets. In particular, the chapters tackle different aspects of how integration has changed over time and the relevance of particular global factors in pricing.
In Chapter 1, I study the link between globalization and asset returns. Here, I provide a comprehensive analysis of the impact of economic and financial globalization on asset return comovements over the past 35 years. The globalization indicators draw a distinction between de jure openness that results from changes in the regulatory environment and de facto or realized openness, as well as between capital market restrictions across different asset classes. Although globalization has trended positively for most of the sample, the global financial crisis and its aftermath have provided new headwinds. Equity, bond, and foreign exchange returns often have different responses to globalization. I generally find weak evidence of comovement measures reacting to globalization and often find other economic factors to be equally or more important determinants.
In Chapter 2, I analyze variance risk in global markets. Innovations in volatility constitute a potentially important asset pricing risk factor that can be easily tested through the return on variance swaps. I characterize the exposure of the returns on three asset classes (equities, bonds and currencies) in all regions of the world to United States based equity variance risk. I explore the implications for global risk premiums and asset return comovements using both developed and emerging markets. I first find that regional portfolios across all three asset classes and practically all countries exhibit negative loadings with respect to the variance risk factor. This exposure is not only statistically but also economically significant representing for most assets we consider around 50% of the global risk premiums implied by a simple three-factor model with global equity, bond, and variance risks. Second, this simple three-factor model also explains a substantive fraction of the comovements between international assets, but the fit is best for international equity correlations and is worse for currency returns and across asset correlations.
In Chapter 3, I study the link between time-varying integration and asset pricing. Emerging markets are subject to constant integration shocks, which can make markets more integrated or more segmented. Changes in integration have dynamic effects that are difficult to accommodate in valuation models, as both time-varying betas and risk premium are needed to capture the direct and indirect effects of changes in integration on dividend yields. Here, I develop a novel present value model to value cash flows with time-varying expected returns, where integration affects the cost of capital in a time-varying fashion. This framework prices expectations about future integration, which is modeled as a mean reverting process. I calibrate the model using a segmentation shock in Argentina in 2011 as a case study, and find that the model is able to capture part of the increase in dividend yields as markets became more segmented. By assuming that investors perceive the shock as permanent and thus price lower mean integration following the segmentation shock, I am able to model the full extent of the change in dividends.
The three chapters show that, while integration has broadly increased over time, different asset classes have different responses to globalization. I find that integration is time-varying and that markets can become more segmented; that is, integration is not a one-way street, as many models have assumed in the past. Finally, I show that global factors matter in emerging markets in all asset classes, and identify variance risk as a new risk factor which helps explain why global capital asset pricing models tend to yield low discount rates in these economies. Therefore, researchers and practitioners should take into account the importance of both local and global factors when valuing emerging market assets and take into account that the relative importance of each factor varies over time.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/d8-13qc-sv68 |
| Date | January 2019 |
| Creators | Kiguel, Andrea |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0021 seconds