Theory Lecture 5 Flashcards
What is the motivation behind International risk sharing (or risk diversification)?
A part of income comes from abroad -> Lower volatility of income without any net lending or borrowing at home.
Describe the theory behind international risk sharing (diversification of risk)
1) Country-specific output risks are pooled in global financial markets, and domestic consumption growth should not depend too heavily on country-specific income shocks.
2) This brings large theoretical welfare gains, Van Wincoop ’94JME; Tesar
’95JIMF.
What are the two assumptions made in the simplest models of international finance?
1) efficient markets (for goods/services, assets, etc.)
2) “representative” rational risk-averse investors/consumers.
What is the issue with the simplest models of international finance?
International macro-finance faced a number of empirical regularities
(“puzzles”) that are at odds with the simplest economical model.
What are the anomalies or “puzzles” in international finance, related to international asset portfolio diversification ?
- Home bias in equity portfolios.
2. Lack of international risk sharing.
What are the finding of Home Bias in Equity Portfolios anomaly ? (what is it, how is it measured)
Measure of EHB is:
The difference between actual holdings of domestic equity and the share of domestic equity in the world market portfolio (assume basic World CAPM)
•Casual evidence:
-> By 2000, foreign equity was only about 12% in the U.S. portfolio – the “optimal share” is 50% (Ahearne, Griever, Warnock ’04JIE).
->Stock-market investors maintain a puzzling preference for home assets.
Portfolio holdings of American households by wealth deciles (Household)
Share of domestic public debt held by banking and financial sectors – a growing exposure of domestic banks to domestic sovereign debt (Banks)
Is Equity Home Bias really an anomaly?
•Lewis 2011:
Whether or not it is a puzzle depends on whether such home-biased portfolios achieve the objectives of home investors.
•Puzzle if the “correct investor model” is a single-factor World CAPM
-> Whatever the “true” model, it is puzzling that investors do not reap the diversification benefits that foreign equities seem to offer.
Definition of risk sharing
The risk sharing the ability of a country to insulate its
consumption from shocks to its own output, …
…after controlling for the component of output growth that is
common across countries.
Can The systematic “global” risk be diversified?
The systematic “global” risk cannot be diversified, as in we know from the portfolio theory in asset pricing
How can one measure risk sharing?
Sorensen & Yosha ’98JIE international extension of Asdrubali et al. ‘96QJE.
•Intuition of the measure: With full risk sharing, consumption growth is independent of idiosyncratic shocks to national output (GDP)
How do we achieve “separation” of consumption fluctuations from fluctuation of output (domestic business cycle)?
Channels of R/S use the decomposition from the System of National Accounts
What is the effect of risk sharing during crisis?
Risk sharing collapsed rapidly during the crisis
Explain the lack of international risk sharing
•….there is little evidence that countries are using global financial markets to smooth
consumption (imperfect risk sharing).
•Lewis ’99JEL: costs to diversification (cross border frictions) > benefits (smoothing of
shocks) involved.
Empirical studies on what frictions matter for R/S include Lewis ’96JPE; Sorensen, Wu, Yosha,
Zhu ’07JIMF; Kose, Prasad, Terrones ’09JDevE; Fratzscher & Imbs ’09JFE.
R/S is larger within countries than between (Crucini ‘99REStat; Kalemli-Ozcan, Sorensen,
Yosha ’03AER; Becker & Hoffmann ’06EER).
•The lack of international risk sharing is related to the Feldstein-Horioka & equity home
bias puzzles.
Empirical research shows that many of the theoretical diversification benefits are not
fully exploited by countries. Which risk sharing could be improved or exploited more?
Part of consumer risk sharing may be obtained by domestic mechanisms (government transfers, etc.). See Asdrubali et al ‘96 for “channels”.