Page 413 - Special Topic Session (STS) - Volume 3

STS552 Carol C. Bertaut et al.
These findings have implications for understanding the factors influencing
capital flows. For example, there has been much focus on the global impact of
the extraordinary policy actions undertaken by advanced economy central
banks in the wake of the global financial crisis. Of particular emphasis has been
how these monetary policies spill over into emerging markets and how EME
asset prices will react when these policies are reversed (Bowman et al 2015,
Fratzscher et al 2018, Curcuru et al 2018). Our results showing understated
growth in holdings of EME assets also imply mismeasurement of capital flows
to EMEs. Overall, flows appear to have been stronger when policy was
especially accommodative, which suggests that the spillovers may be
understated.
Our results also weaken the argument that capital flows arising from foreign
direct investment (FDI) are generally preferable because they are less volatile
than portfolio flows, in part because FDI is harder to expropriate (Albuquerque
2003) and is driven by pull rather than push factors (Eichengreen et al 2018).
However, these arguments assume that portfolio flows in the BOP accounts fully
capture investment in a country’s securities. When foreign residents buy bonds
issued onshore, these purchases will show up as portfolio investment inflows.
When corporations issue bonds via offshore affiliates, however, funds borrowed
through the offshore entities are funnelled back to the parent firm in the form
of lending or “reverse investment” in the parent firm. These flows, which will
appear as FDI inflows, are effectively no different from typical portfolio flows,
and can be just as volatile. Growing reliance on offshore financing vehicles for
debt issuance can thus confound our understanding of the resilience of different
types of cross-border financial flows. Similarly, our results also raise some
potential flags for interpreting conclusions on the effectiveness of capital
controls in preventing portfolio inflows to emerging markets (Forbes and
Warnock 2012; Ahmed and Zlate 2014; Forbes et al. 2014; Forbes et al 2015;
Pasricha et al. 2015). Foreign investors may still be able to gain exposures to
countries via offshore-issued bonds, which typically are unaffected by controls.
But because such purchases are not classified as portfolio inflows to these
countries, the effectiveness of the controls may be overstated.
Our results are also relevant to the long-standing Lucas (1990) paradox,
which arises from differences between the theoretical prediction of
movements between developed and developing countries, and what is
observed. Theory predicts that capital should move toward economies with
lower levels of capital per worker. Contrary to this theory, most studies find
that capital does not flow from more to less developed economies; rather, it
flows in the other direction (see Alfaro et al. 2008, among others). Our results
suggest that advanced economy exposure to EMEs is larger than previously
believed, which resolves some portion of this puzzle. This is perhaps especially
evident when we consider the global reaches of large multinational firms. In

402 | I S I W S C 2 0 1 9