Assessing the Impacts of Global Financial Integration

近十年来，国际金融市场一体化的趋势加快了。全球市场一体化对资产价格的影响显然是一个重要的政策和研究问题。股票价格是更多地作为一种资产在一个单一的世界市场定价，还是仍然根据本国的基准定价？如果股票市场正在演变成一个共同的一体化市场，那么关于外国资产多样化潜力的传统假设对资产价格的影响意味着什么？这些一般性问题是宏观经济和金融研究领域大量文献的焦点。然而，这个项目打算采取一种不同的方法来解决这些问题。人们普遍认为，外国股票的高成本会干扰全球市场一体化。这些成本通常被认为是两种类型之一：(1)进入国外市场的直接交易成本，并支付可能更高的税费；或者(2)了解外国公司及其成长潜力的信息成本。本项目考察了一组外国股票，这些股票在这两个方面基本上没有明显更高的成本。调查人员建立了一个独特的数据集，其中包含通过所谓的ADR（美国存托凭证）计划在美国交叉上市的外国公司的信息。购买这些外国股票的成本并不比购买国内股票高。此外，在纽交所上市这些股票的外国公司必须遵循与美国本土公司相同的披露要求，包括遵循相同的会计准则。因此，国内市场上的这些外国股票提供了一个独特的机会，可以直接研究全球市场一体化对资产定价的影响，以及对国际多元化收益潜力的影响。该数据集是独一无二的，因为它提供了外国公司在国内和美国的价格的完整时间序列。该项目的第一部分侧重于金融一体化对资产定价关系的影响。国外股票在美国市场的发行与一个偶然的实证观察有关：母公司交叉上市后的股价与美国市场的协方差趋于更正。如果这些轶事是真实的，它们对理解国际股票定价很重要，从而对大多数与国际金融市场有关的政策问题产生影响。该项目的这一部分研究的问题是，以股票交叉上市的形式进行的更大的国际金融一体化是否影响了国际股票定价。它还计算了整合前后国际股票市场的福利收益。该项目的第二部分更直接地侧重于面对更大一体化的多样化范围。这个问题对于从退休及其相关的社会保障问题到国际资本分配问题等政策问题都很重要。该项目使用了大约2000家外国公司在美国交叉上市的股票价格的时间序列，并将其与本国市场的股票价格进行比对。这样就可以比较交叉上市前后的资产定价行为。使用贝叶斯方法可以计算出一旦这些股票在美国发行，投资者会选择什么，从而解决了两个不同的基本问题。首先，比较交叉上市前后美国与某一外国市场的整合程度，整合程度提高带来的多元化收益是什么？这个问题很重要，因为现有文献一般只比较了相对于完美整合的收益。第二，持有交叉上市外国股票的美国投资者还会选择增持非交叉上市外国股票吗？显然，回答这个问题对于确定国内投资者是否真的需要持有直接从国外证券交易所获得的资产以实现最佳多元化很重要。第三部分的问题是：通过交叉上市的全球一体化是否提高了国内居民对冲消费风险的能力？它通过检验由这些股票解释的消费增长的可变性是否显著高于一般外国指数的可变性来解决这个问题。处理这些问题是重要的，因为从宏观经济角度对国际多样化的研究侧重于分担消费风险。在考虑国际投资组合配置的程度上，它一直是在问，哪些资产将支持消费风险的分担。迄今为止，尚无研究考察通过交叉上市的市场整合对消费风险保险能力的直接影响。

The trend toward international financial market integration has accelerated over the past decade. The effects of global market integration on asset prices is obviously an important issue for both policy-making and research reasons. Are equity prices being priced more as an asset in a single world market or are they still priced according to benchmarks from their own domestic country? And if equity markets are evolving towards a common integrated market, what do the effects on asset prices imply about traditional assumptions about the diversification potential of foreign assets? These general questions are the focus of a large literature spanning both macroeconomic and financial research. This project intends to take a different approach to these questions, however. It is commonly argued that the higher costs of foreign stocks interfere with global market integration. These costs are usually argued to be one of two kinds: (1) direct transactions costs of going to foreign markets and paying possibly higher taxes and fees; or (2) informational costs of learning about a foreign firm and its growth potential. This project examines a set of foreign stocks that are largely devoid of significantly higher costs along these two dimensions. The investigator builds a unique data set with information about foreign companies that cross-list their stocks in the U.S. through so-called ADR (American Depositary Receipt) programs. Acquiring these foreign stocks is no more costly than acquiring domestic stocks. Moreover, foreign firms that list these stocks on the NYSE must go through the same disclosure requirements as a domestic US firm, including following the same accounting standards. Therefore, these foreign stocks on domestic markets provide a unique opportunity to directly study the effects of global market integration on asset pricing and on the potentials for gains from international diversification. The data set is unique because it provides a full time series for the price of the foreign companies both at home and in the U.S. This project consists of three parts. The first part of the project focuses upon the effects of financial integration on asset pricing relationships. The issue of foreign stocks in the US market has been associated with a casual empirical observation: the stock price of the parent company tends to covary more positively with the US market after cross-listing. If these anecdotal stories are true, they are important for understanding international equity pricing, and thereby bear on most policy issues concerning international financial markets. This part of the project examines the question of whether greater international financial integration in the form of cross-listings of stocks has affected international equity pricing. It also calculates the welfare gains for international equity markets before and after integration. The second part of the project focuses more directly on the scope for diversification in the face of greater integration. This question is important for policy issues ranging from retirement and its associated social security questions to international allocation of capital issues. The project uses a time series of about 2000 foreign firms' cross-listed stock price in the US and match these with their counterparts on their own markets. This allows a comparison of asset pricing behavior before and after cross-listing. A Bayesian approach is used that allows the calculation of what investors would choose once these stocks are issued in the U.S. and thereby address two different fundamental questions. First, comparing the degree of integration between the US and a given foreign market before and after cross-listing, what are the diversification gains from the increased integration? This question is important because the existing literature has generally compared gains only relative to perfect integration. Second, would US investors who hold cross-listed foreign stocks still choose to hold additional foreign stocks that are not cross-listed? Obviously, answering this question is important for determining whether domestic investors really need to hold assets obtained directly from foreign stock exchanges to be optimally diversified. The third part of the project asks: Has global integration through cross-listing improved the domestic resident's ability to hedge consumption risk? It addresses this issue by testing whether the variability of consumption growth that is explained by these stocks is significantly higher than the variability of the general foreign index. Addressing these issues is important since studies of international diversification from a macroeconomic viewpoint have focused upon the sharing of consumption risks. To the extent that international portfolio allocation has been considered, it has been to ask what assets would support the sharing of consumption risks. No study to date has examined the direct effects of market integration through cross-listing on the ability to insure consumption risk.