Periodic Reporting for period 1 - GlobalPolicyUncertainty (Global Policy Uncertainty and International Asset Markets)
Reporting period: 2016-09-22 to 2018-09-21
Governments establish policies and regulations that impact on economic growth and welfare. A ubiquitous byproduct of policymaking is policy uncertainty, which may have severe adverse effects on the economy, for instance, leading to lower investment, reduced trade activity and lower asset prices. Moreover, given the interconnectedness of the world economy, policy uncertainty has spillover effects beyond national borders. Recent events, such as the US elections, Brexit, or the budget dispute in Italy, have been associated with an increase in policy uncertainty within and across borders, which pose challenging questions for the portfolio construction of international investors. In times of high policy uncertainty, investors commonly seek shelter in safe haven assets such as sovereign bonds. Consequently, both domestic and foreign policy uncertainty affect government bond yields.
The main objective of the project is to understand how a country’s yield curve responds to a change in both domestic and foreign policy uncertainty. In light of the high policy uncertainty of recent years and its profound impact on financial markets, we argue that a deeper understanding of the effects of economic policy uncertainty on global yield curve is both timely and relevant. The novelty of the project lies in the fact that, contrarily to studies that analyzed the impact of policy uncertainty within a country, this project contributes by understanding government policy uncertainty as a global phenomenon and studying its impact on international financial markets. The project links several strands of economic literature, such as international finance, empirical asset pricing, and macroeconomics, and contributes to an active and growing literature on policy uncertainty by exploring its international dimensions
The main objective of the project is to understand how a country’s yield curve responds to a change in both domestic and foreign policy uncertainty. In light of the high policy uncertainty of recent years and its profound impact on financial markets, we argue that a deeper understanding of the effects of economic policy uncertainty on global yield curve is both timely and relevant. The novelty of the project lies in the fact that, contrarily to studies that analyzed the impact of policy uncertainty within a country, this project contributes by understanding government policy uncertainty as a global phenomenon and studying its impact on international financial markets. The project links several strands of economic literature, such as international finance, empirical asset pricing, and macroeconomics, and contributes to an active and growing literature on policy uncertainty by exploring its international dimensions
To analyze the relationship between global uncertainty and sovereign bond markets, we construct a monthly panel dataset for 19 countries from January 1990 to June 2018. As a measure of uncertainty, we consider the Economic Policy Uncertainty Index (EPU) (Baker, Bloom and Davis (2016)), which reflects uncertainty about fiscal, regulatory or monetary policy and it is constructed using the relative frequency of own-country newspaper articles that contain the trio of terms: economy (E), policy (P) and uncertainty (U). As financial indicators we consider the zero-coupon yield curve for government bonds with maturities 1 - 10 years.
In the first part of the analysis, we decompose each country’s EPU index into global and local (country-specific) uncertainty factors. The global factor is constructed as the GDP-weighted average of national EPU, and the local factor is obtained as a residual that reflects the domestic uncertainty that cannot be explained through the global factor. We find that countries differ in their exposure to the global factor and in the volatility of their local uncertainty. With these two measures uncertainty, in the second part of the analysis we study their differential effects on the government yield curve. We find that both sources of uncertainty are important in determining government yields; however, yields respond very differently to global and local sources of policy uncertainty.
Figure 1 shows a robust negative relationship between government yields and global uncertainty at all maturities, which is consistent with the response to an undiversifiable risk: when global uncertainty is higher, capitals fly away from risky investments towards safe assets, pushing yields down in all countries. In contrast, we find an heterogeneous response of government yields to local uncertainty (both negative and positive ‘betas’, which are more dispersed for longer maturities). This result suggests that, in some countries, higher local uncertainty shifts capital towards their own governments bonds, whereas in other countries, capital flies across the border towards foreign bonds. This heterogeneous cross-country flight-to-safety phenomenon is a novel result, and it is robust to including additional measures of global risks and global recessions.
In order to explain the heterogeneous responses to local uncertainty, we correlate the response of government yields with different economic, financial, and institutional characteristics of a country, such as financing costs, bond market liquidity, financial depth, and governance quality. We discover that these characteristics are important determinants of the heterogeneous responses to local uncertainty.
Figure 2 shows that cheaper financing costs, higher liquidity, stronger financial depth and better governance favour the role of a country’s government bonds as a safe-haven against local uncertainty.
The project has been presented at universities, central banks and other research institutions. I organized the Workshop “Uncertainty in Macroeconomics and Finance” where a large group of researchers discussed about the effects of uncertainty in global financial markets and trade networks, asset markets, labour markets, monetary and fiscal policy, as well as advancements in the measurement of uncertainty at national and international levels.
In the first part of the analysis, we decompose each country’s EPU index into global and local (country-specific) uncertainty factors. The global factor is constructed as the GDP-weighted average of national EPU, and the local factor is obtained as a residual that reflects the domestic uncertainty that cannot be explained through the global factor. We find that countries differ in their exposure to the global factor and in the volatility of their local uncertainty. With these two measures uncertainty, in the second part of the analysis we study their differential effects on the government yield curve. We find that both sources of uncertainty are important in determining government yields; however, yields respond very differently to global and local sources of policy uncertainty.
Figure 1 shows a robust negative relationship between government yields and global uncertainty at all maturities, which is consistent with the response to an undiversifiable risk: when global uncertainty is higher, capitals fly away from risky investments towards safe assets, pushing yields down in all countries. In contrast, we find an heterogeneous response of government yields to local uncertainty (both negative and positive ‘betas’, which are more dispersed for longer maturities). This result suggests that, in some countries, higher local uncertainty shifts capital towards their own governments bonds, whereas in other countries, capital flies across the border towards foreign bonds. This heterogeneous cross-country flight-to-safety phenomenon is a novel result, and it is robust to including additional measures of global risks and global recessions.
In order to explain the heterogeneous responses to local uncertainty, we correlate the response of government yields with different economic, financial, and institutional characteristics of a country, such as financing costs, bond market liquidity, financial depth, and governance quality. We discover that these characteristics are important determinants of the heterogeneous responses to local uncertainty.
Figure 2 shows that cheaper financing costs, higher liquidity, stronger financial depth and better governance favour the role of a country’s government bonds as a safe-haven against local uncertainty.
The project has been presented at universities, central banks and other research institutions. I organized the Workshop “Uncertainty in Macroeconomics and Finance” where a large group of researchers discussed about the effects of uncertainty in global financial markets and trade networks, asset markets, labour markets, monetary and fiscal policy, as well as advancements in the measurement of uncertainty at national and international levels.
One promising future project consists in applying our empirical strategy to analyze the effects of global policy uncertainty to a broader set of asset classes at different maturities, i.e. corporate bonds, stocks, exchange rates and derivatives. For this purpose, a panel of financial instruments and asset prices must be put together; the advantage is that data on economic policy uncertainty, its decomposition into global and local factors and other tools developed for the project with government bonds are available.
A second future project consists of a theoretical framework that can be used to discipline the empirical analysis, run comparative statics, predict outcomes, and evaluate policy recommendations. This model will build upon the framework developed in a companion paper, “Might Global Uncertainty Promote International Trade?” where we develop a framework to think about the role of global uncertainty in shaping international trade in goods markets. Such a model will be expanded to incorporate investment opportunities in government bonds and other asset classes across countries. In this multi-country general equilibrium asset-pricing model, government policy uncertainty (both global and country specific) becomes a source of priced risk in international stock and bond markets.
Within this model, we will characterize the effects of uncertainty on the yield curve, returns’ volatilities and correlations, among other characteristics of the financial instruments. Secondly, we will conduct comparative static analysis to assess how other variables affect the relative importance of local vis-à-vis global uncertainty Then, we will use the model to quantify how financial integration, the level of asymmetries in information, countries’ importance in global risks, and other variables determine the spillovers of uncertainty across countries and the sensitivity of bond yields and stocks prices to uncertainty. In particular, we expect global uncertainty to be more important as a pricing factor as the level of financial integration across countries increases. Finally, we expect to be able to assess the impact of uncertainty-reducing welfare-improving strategies such as increasing government transparency and improving communication about policy.
A second future project consists of a theoretical framework that can be used to discipline the empirical analysis, run comparative statics, predict outcomes, and evaluate policy recommendations. This model will build upon the framework developed in a companion paper, “Might Global Uncertainty Promote International Trade?” where we develop a framework to think about the role of global uncertainty in shaping international trade in goods markets. Such a model will be expanded to incorporate investment opportunities in government bonds and other asset classes across countries. In this multi-country general equilibrium asset-pricing model, government policy uncertainty (both global and country specific) becomes a source of priced risk in international stock and bond markets.
Within this model, we will characterize the effects of uncertainty on the yield curve, returns’ volatilities and correlations, among other characteristics of the financial instruments. Secondly, we will conduct comparative static analysis to assess how other variables affect the relative importance of local vis-à-vis global uncertainty Then, we will use the model to quantify how financial integration, the level of asymmetries in information, countries’ importance in global risks, and other variables determine the spillovers of uncertainty across countries and the sensitivity of bond yields and stocks prices to uncertainty. In particular, we expect global uncertainty to be more important as a pricing factor as the level of financial integration across countries increases. Finally, we expect to be able to assess the impact of uncertainty-reducing welfare-improving strategies such as increasing government transparency and improving communication about policy.