Daily data for such a large number of countries are not available (Yan 2015).
These are the main markets covered by the previous literature.
The sample size of twelve markets is large enough to provide
results that are potentially fairly general, yet it is small enough
to allow more attention to market-specific analysis and presenting results market-by-market in an intelligible way, which
may be more difficult using a larger number of markets. My
sample markets have been studied in earlier literature. For
example, Richards (2005) looked at Indonesia, South Korea, the
Philippines, Thailand, and Taiwan Republic of China. Fuertes
et al. (2016) and Yan et al. (2016) include all these markets as a
subsample of their studies. These markets are of vital importance in the global economy in terms of gross domestic product
and the amounts of capital flows.
I scale the observed equity flows by the U.S. consumer price
index (CPI) to eliminate the impact of inflation effects. I take
the price in 1993 as the price of the base year in the United
States. My results do not qualitatively change when I repeat the
analysis process based on the un-scaled short-term equity
flows data. CPI data are obtained from Datastream.
Following Sarno and Taylor (1999a,b) and Fuertes et al. (2016),
I decompose the observed equity flows from the United States
to twelve EMs into unobserved permanent and temporary components and identify hot money in equity flows as the temporary
component via deploying state-space models using a Kalman
exchange, Yan (2015) exploits the interaction between equity
flows and stock returns and provides some new evidence on
foreign investors’ trading behavior and their price impact, and
finds that the bi-directional causality is plausible; that is, equity
flows have a positive impact on equity returns and vice versa.
Although none of the previous papers studies whether and how
local equity markets drive the hot money, recent literature (e.g.,
Fuertes et al. 2016) provides me an opportunity to do so.
I collect monthly bilateral capital outflow and inflow data in
millions of U.S. dollars during January 1993–December 2013
from the U.S. Treasury International Capital database. “Gross
purchases by foreigners” and “gross sales by foreigners” are
classified as U.S. sales and U.S. purchases, respectively, in the
International Capital Reports of the U.S. Treasury Department.
The data are collected and presented from the perspective of
the foreign parties to the transactions. By definition, “gross
purchases by foreigners” are gross sales by U.S. residents.
Similarly, “gross sales by foreigners” are gross purchases by
U.S. residents. A positive difference indicates net foreign
purchases from U.S. residents (U.S. capital inflow) and a
negative difference indicates net foreign sales to U.S. residents
(U.S. capital outflow).
The twelve EMs in my sample are Argentina, Brazil, People’s
Republic of China, Chile, Indonesia, India, South Korea,
Mexico, Malaysia, the Philippines, Thailand, and Taiwan
Republic of China. There are eight Asian markets: People’s
Republic of China, Indonesia, India, South Korea, Malaysia, the
Philippines, Thailand, and Taiwan Republic of China: and four
table 1 reports the results from state-space models for net equity flows, which are cPi-scaled capital flows in millions of U.S. dollars.
A dash indicates that the component at hand is absent from the model. 0 ≤ q-ratio ≤ 1 is the standard deviation of each component
over the largest standard deviation across components, computed from the variance-covariance matrix of disturbances. Column seven reports the final level of the stochastic trend and its root mean square error (RMSE). The last column reports the R2. The sampling
frequency is monthly during January 1993–december 2013. the country abbreviations are listed as follows: Argentina (Ag), Brazil (BR),
People’s Republic of China (CH), Chile (CL), Indonesia (ID), India (IN), South Korea (KO), Mexico (MX), Malaysia (MY), Philippines (PH),
Thailand (TH), and Taiwan Republic of China (TW).
Final level of stochastic
trend [RMSE] R2
AG 0.000 0.202 1.000 1.754 –0.758 −0.655[0.226] 0.492
BR 0.000 0.809 1.000 0.774 – 3.487[0.026] 0.353
CH 0.000 0.569 1.000 0.688 – −0.784[0.355] 0.404
CL 0.000 0.249 1.000 0.683 −0.116 −0.471[0.097] 0.467
ID 0.000 0.514 1.000 0.672 – 0.203[0.184] 0.391
IN 0.000 0.520 1.000 0.858 – 1.392[0.057] 0.423
KO 0.000 0.789 1.000 0.892 – 0.482[0.669] 0.372
MX 0.000 0.894 1.000 0.841 −0.177 − 1.582[0.009] 0.353
MY 0.000 − 1.000 – – −0.067[0.511] 0.334
PH 0.000 0.480 1.000 0.848 – −0.034[0.586] 0.404
TH 0.000 0.486 1.000 0.715 – 0.260[0.008] 0.397
TW 0.000 1.000 0.000 0.510 – 0.9780[0.470] 0.255