While consistent with both survey evidence and previous
empirical research, the natural hedging motive appears
to be only a partial explanation on examination of
the actual data on FC debt issuance. As noted above, existing evidence for natural hedging-based FC debt issuance
comes from studies of nonfinancial firms. These
firms, however, collectively represent only a very small
share of FC debt issuance. The vast majority of FC debt is
issued by financial firms and sovereign entities. A recent
study of FC debt issuance by Munro and Wooldridge
(2009), for example, finds that financial firms represent
the single largest category of FC debt issuers, with government
issuers representing the second largest. Nonfinancial
issuers, by contrast, represent less than 10% of
the FC bonds issued during the 2000s. A study by Habib
and Joy (2008) on a similarly comprehensive data set of
FC bonds issued between 1999 and 2008 finds that financial
firms represent between 50% and 80% of the FC
bonds with 5-year maturities issued in the five major international
currencies. They represent an even larger
share of the FC bonds issued with shorter and longer
maturities.
Government and other sovereign entities obviously
have very limited foreign currency cash flows to hedge.
Banks and other financial firms in a given country, on the
other hand, may well make a large number of foreign
loans or hold a significant amount of foreign assets that
they wish to hedge (or finance) with FC debt. It seems
unlikely, however, that the overall currency composition
of these foreign loans and securities portfolios would
vary significantly from year to year. By contrast, the relative
share of total FC debt that is issued in each major
international currency fluctuates widely. As a case in
point, in the study by Habib and Joy, the euro’s share
of FC bond issuance varies between 19.1% of all FC
debt issues in 2002 and 40.2% in 2005. Similarly, the
US dollar’s share ranges from 60% of all FC debt issues
in 2001 to less than 33% in 2005. If bank-lending patterns
and foreign securities portfolios remain relatively
constant over time, it is challenging to reconcile this pattern
of issuance with the classic natural hedging motive.
To illustrate these points over an even longer sample
period, Table 15.1 presents data on a comprehensive
sample of FC bonds issued from 1991 to 2010. Like previous
studies, in order to zero in on FC debt only, the
sample excludes bonds denominated in the home currency
of each issuer. Since many emerging market nations
included in the sample pegged their currencies to
the US dollar over a portion of the sample period, the
sample excludes US dollar-denominated bonds for issuers
in these countries during the relevant time periods
as well. The resulting sample has 37 357 bond offerings in
71 different currencies from issuers domiciled in 101 different
countries. Panel A of Table 15.1 presents the total
number of bonds issued in each of the top 10 international
currencies by issuers domiciled in each of the
top 25 most active issuer nations throughout the entire
sample period.
Panel B of Table 15.1 confirms the results of previous
studies with regard to the relative significance of financial
firms and government entities in terms of overall FC
debt issuance. Consistent with previous findings, financial
firms represent the 60% of all FC debt issuance, government
entities represent 22%, and industrial firms and
all other issuers represent only 17%. Panel C of Table 15.1
similarly confirms previous results with regard to the extreme
variation across time in the relative share of total
FC debt that is issued in each major currency. As evident
in the table, the share of bonds issued in each of the top
10 international currencies varies considerably throughout
the sample period. The Japanese yen’s share is an extreme
case in point. It increases from 5% in 1992 to nearly
37% in 1995 and 1996 before falling back to 5% by 2003.
Similar fluctuations in relative currency share are