Sebnem Kalemli-Ozcan and Elias Papaioannou explain that when shocks are primarily in productivity, a higher degree of banking integration results in a lower degree of business cycle synchronization, but when banks suffer an exogenous shortfall in their revenues, economic activity compresses in financially integrated countries through higher interest rates.
While many commentators, academics, and policy makers argue that financial globalization has been a catalyst for the propagation of the 2007-2009 crisis from a corner of the US capital markets to the rest of the world, so far the tentative empirical evidence is inconclusive. Studies fail to find that countries with stronger financial linkages to the US experienced sharper recessions as compared to less inter-connected economies. In contrast to this lack of evidence, most theoretical models in international macroeconomics and corporate finance suggest that by facilitating contagion, capital supply shocks will lead to more synchronized cycles among financially integrated economies.
A key problem is that even before the recent crisis we lacked a good understanding of how financial integration affects the synchronization of economic activity in regular times. Empirical works show a clear positive association between international financial integration and output synchronization across pairs of industrial and emerging countries in the 1990s and early 2000s. Yet this result is at odds with standard theory predicting that, in the absence of major financial shocks, financial globalization and the synchronization of economic activity should be negatively correlated. In the textbook international business cycle model, following a positive country-specific productivity shock, the return to capital and labor increases and foreign capital flows to finance the rising investment opportunities; consequently output patterns among financially integrated countries diverge.
To make the long story short, there is a paradox between theoretical predictions and the main data patterns both in tranquil times and during the 2007/2009 global financial crisis. While in regular times the partial correlation between financial integration and business cycle synchronization should be negative, there is a positive association in the data both across country pairs and across time. Furthermore, there is no evidence on the recent crisis that shows a systematic relationship between links in the US capital markets and the spread of the crisis.
1. Identifying the Causal Effect of Financial Integration on Output Synchronization in Tranquil Times
In our paper “Financial Regulation, Financial Globalization and the Synchronization of Economic Activity”, we attempt to identify the one-way effect of financial integration on international business cycle synchronization in regular (non crisis) times. For our analysis we exploit a confidential dataset from the Bank of International Settlements (BIS) that covers all international bilateral banking activities for the twenty largest economies over the past three decades. As we want to study the effect of financial integration on output synchronization in regular times, we focus on the pre-crisis period (1978-2006).
The rich structure of our data is essential as it allows us to account econometrically for many sources of bias. The three-dimensional structure (that records the exposure of banks located in country i to country j in year-quarter t) of our dataset enables to control for global trends and common-to-all-country shocks, such as the increased co-ordination of monetary policy, out-sourcing, and other features of globalization. Quite importantly we also account for time-invariant country-pair factors, related to trust, distance, and information asymmetries that affect both financial integration and the synchronicity of output fluctuations. Besides these straightforward technical merits, by exploiting the within country-pair variation of the data we can address directly the relevant policy question: do increases in bilateral financial linkages makes economic activity more or less synchronized?
1.1. Correlation Analysis
In the first part of our analysis we show with simple econometric methods that it is indeed fundamental to account for common-to-all-country shocks and country-pair fixed-factors. Across country-pairs there is a significant positive correlation between financial integration and output synchronization; this comes at no surprise. The business cycle of the US economy is both more synchronized and more financially linked with Canada, as compared to Germany or France, which are themselves more synchronized and also more interconnected. Yet, in sharp contrast to the positive cross-sectional correlation, when we examine the within country-pair response of output synchronization on increases in bilateral financial linkages we find a significantly negative association. This implies that increases in financial integration within country-pairs (say Canada-US or France-Germany) are associated on average with less synchronized, more divergent, output cycles. The negative within country-pair association between bilateral financial linkages and business cycle co-movement is in line with the standard textbook models in international macro implying that in the absence of financial sector shocks, cross-border financial integration should magnify total-factor-productivity shocks and make output patterns diverge.
1.2. Regulatory Harmonization in Financial Services and Output Synchronicity
While the supervisory BIS data reflect more than 99% of the international exposure of the local banking system, they do not capture other forms of international investment, such as foreign direct investment and portfolio investment between non-banks. Moreover, the BIS data (as most international capital data) miss-record investment channeled via off-shore financial centers. To account for these caveats, we construct an index of financial integration, which is based on the adoption timing of financial sector legislation that aims to harmonize the regulatory framework in financial intermediation across the European Union (EU). Compared to outcome-based indicators (such as international capital holdings and cross-border equity return correlations), employing a time-varying de-jure measure of financial integration allows us to account for reverse causation arising from the fact that international banking may react to the synchronization of output fluctuations, while at the same time accounting for country-pair heterogeneity, global shocks, and common trends.
To construct this index, we exploit the peculiar nature of adopting EU-wide legislation across EU member countries — the EU Directives transposition system. The Financial Services Action Plan (FSAP) was a package of financial reforms launched by the EU in 1998 aiming to integrate the segmented EU financial markets and reduce the costs of cross-border financial intermediation. The FSAP included 29 major pieces of legislation (27 EU Directives and 2 EU Regulations) in banking, capital markets, corporate law, payment systems, and corporate governance. Examples include the Directive on money laundering, the Directive on financial collateral arrangements, the Directive on prospectuses, and the Directive on insider trading and market manipulation. In contrast to EU Regulations that become immediately enforceable across EU member countries, EU Directives are acts that become enforceable only after each EU member country passes domestic legislation adopting explicitly the EU Directive. The legal adoption of the EU Directive (the so-called “transposition” process) is notoriously slow, since it requires modifications of existing institutional structures, the removal of previous regulations and, in many cases, the establishment of new agencies and infrastructure. The transposition of the EU Directives takes in practice several years and differs considerably across EU member states. Using information from the EU Commission on the adoption timing of each of the Directives of the FSAP across EU countries, we construct a bilateral time-varying index that reflects how similar are the legal/regulatory structures governing the functioning of financial intermediation across each country-pair in each year. We then show that a higher degree of legislative/regulatory harmonization in financial services is associated with less synchronized output cycles.
1.3. Estimating the Causal Effect of Financial Integration on Output Synchronization in Regular Times
To identify the causal effect of financial integration on international business cycle synchronization in tranquil times, we develop a novel country-pair panel instrumental variables identification scheme that links legal-regulatory harmonization reforms in financial services with bilateral banking activities in a first-stage empirical model and in turn (in the second stage) with output synchronization. This approach has some nice features. From an econometric standpoint, it accounts for all sorts of biases arising from reverse causation, omitted-variables bias, and measurement error. From a policy standpoint, as many countries are currently in a process to redesign the regulatory framework of financial intermediation, our two stage empirical framework enables us to understand how such reforms may affect output synchronization. This identification strategy is appealing as it links regulatory-legislative reforms in financial intermediation with outcomes (banking integration) in exactly the same sector of the economy and in turn to international output synchronization. The exogeneity assumption for instrument validity is plausible because legislative reforms are at the country-level, while the outcomes we study are bilateral. The exclusivity assumption for instrument validity is also credible because harmonization policies in financial intermediation should affect the synchronization of economic activity primarily by altering cross-border financial activities. As the FSAP was initiated, designed, and implemented with the explicit goal to integrate capital markets among EU member countries, it is quite reasonable that (conditional on other country-pair time-varying factors) it should affect output synchronization by spurring financial integration.
The first-stage specifications reveal that cross-border banking activities increase significantly when countries homogenize the rules governing the function of financial intermediation. The first-stage relationship is strong, even when we condition on the flexibility of the exchange rate regime (i.e. that captures the direct effect of the euro), trade, and other (country-pair time-varying) factors. This suggests that a considerable part of the overall positive effect of the single currency in spurring financial integration in Europe comes from regulatory-legislative convergence in financial intermediation.2 The second-stage estimates show that, conditional on common-to-all-country shocks and country-pair time-invariant factors, the exogenous component of banking integration predicted by legislative-regulatory harmonization policies in financial intermediation tends to make business cycles less alike.
Our empirical results therefore suggest that financial globalization has led to more divergent output cycles. Yet this negative association was masked, because over the past two-and-a-half decades the spur in financial globalization coincided with an increased degree of business cycles convergence.
2. Financial Integration and Output Synchronization in Crisis Times
Following the 2007-2008 meltdown in U.S. capital markets, we have observed a massive and contemporaneous drop in output across most major developed countries. Yet in spite of this overwhelming synchronization of the economic activity that dwarfs anything in comparison since the 1970s, most recent empirical studies fail to detect a systemic association between financial linkages to the U.S. with output decline during the past years. This has led many academics and policy makers to argue that the output collapse that developed economies experienced in the past years was the product of a common shock, either in financial intermediation or in firms’ productivity. In recent work with Fabrizio Perri, we investigate theoretically and empirically whether the 2007-2009 global recession was the outcome of a U.S. idiosyncratic shock in financial intermediation that spread contagiously across the world via international financial linkages or whether in contrast the recession and the associated synchronicity of economic activity was the outcome of a common global shock.
2.1. Data Patterns. Financial Linkages and Business Synchronization during the 2007/2009 and Past Financial Crises
We start our analysis examining the partial effect (correlation) of cross-border financial integration, as reflected in BIS data on the international exposure of the banking system of advanced countries, on output synchronization both during tranquil times and during the recent financial crisis. Our analysis using quarterly data from 18 industrial countries from the late 1970s till the end of 2010 reveals some noteworthy patterns.
First, using simple econometric models, we verify our previous evidence that before the recent 2007-2009 financial crisis the within country-pair correlation between cross-border financial integration and output synchronization is significantly negative.
Second, when we allow the effect of financial integration on output synchronization to differ after mid 2007 when the first troubles in the US mortgage markets emerged and banks started suffering loses in their balance sheets, we find that the partial effect of financial integration on output synchronicity turns positive.
Third, we obtain a similar positive partial correlation between financial integration and output synchronization during previous large financial crisis episodes in other developed countries, such as Finland and Sweden in the early 1990s and Japan in the mid/late 1990s. These results suggest that while the recent financial crisis had clearly larger and more severe repercussions for the global economy, the transmission of the shocks did not differ much, as compared to previous (somewhat comparable) banking crises in advanced countries.
Fourth, we examine in detail whether linkages to the US financial system correlate with output synchronicity during the recent financial crisis. In contrast to most previous works, we utilize the richness of the BIS dataset that also records financial transactions between industrial countries and some important off-shore centers (such as the Cayman Islands) and examine the effect of both direct and indirect exposures via small off-shore centers exposure to the U.S. financial system. To deal with indirect exposure to the U.S. via financial centers, we construct a lower and upper bound for the exposure to the U.S. As a lower bound we use direct banking linkages between each country-pair and the U.S. As an upper bound we add exposure to the direct exposure linkages to the Cayman Islands. We find that the positive correlation between output synchronization and exposure to the U.S. financial system—that supports the contagion via financial linkages explanation of the recent crisis—emerges only when, on top of direct links to the U.S., we also consider indirect links via the Cayman Islands and other financial centers. This result suggests that policy makers (and academics) should also examine carefully the implications of the rising important of small financial centers, such as Panama, the Channel Islands, Bermuda, in the functioning of international capital markets.
2.2. An Integrated Theoretical Framework
After establishing the main data patterns, we develop a dynamic stochastic general equilibrium model of international banking that reconciles our empirical results and helps understand the underlying sources of output fluctuations in advanced countries over the past three decades.
Our set-up allows for shocks both to the productivity of firms and to their liquidity (working capital, provided by domestic and international banks); countries differ in the degree of cross-border financial integration, which is reflected in the degree of domestic banks international lending (to foreign firms). The model predicts that when the underlying shocks are primarily in the productivity side, then a higher degree of banking integration results in a lower degree of business cycle synchronization. The mechanism is standard. When a country is hit by a positive productivity shock, a higher degree of financial integration enables capital to flow from the less to the more productive country; as the return of both labor and capital increases, workers work more and as such employment, investment and output diverge further. In contrast when banks suffer an exogenous shortfall in their revenues, they are forced to make up for the lost revenues charging higher interest rates to firms; this, through a working capital (liquidity) channel compresses economic activity in both financially integrated countries. Thus when liquidity (capital supply) shocks in the banking sector dominate, then a higher degree of banking integration results in higher output synchronization, since the interest rates charged by global banks increases. Using simulated data from our model, we run a similar regression as the one we run with the actual data and find similar results: in times with dominant productivity shocks, the relation between integration and synchronization is negative and roughly of the same magnitude as the one estimated on the data in the period 1978-2006; also we find that the coefficient on integration interacted with periods with large credit shocks is positive and again of similar magnitude as the one estimated on the periods of financial crisis in the data.
Sebnem Kalemli-Ozcan is Professor of Economics at University of Houston and a Visiting Professor of Economics at Koc University and Visiting Professor of Public Policy at Harvard’s Kennedy School of Government. Elias Papaioannou is Assistant Professor of Economics at Dartmouth College and a Visiting Assistant Professor at the Department of Economics at Harvard University. This article draws on their recent research with Jose-Luis Peydro and Fabrizio Perri.