Page 329 - Contributed Paper Session (CPS) - Volume 7

CPS2118 Yang Wang
The latest SNA (2008) confirms the importance of these two approaches:
C  I  G X M
GDP  t t t  t  t
o
t
P P x P m
t t t (1)
C  I  G X  M
GDP  e t t t  t t
t *
P t P t (2)
As SNA (2008) stated, there is no consensus on the choice of P*. Kohli
(2004) suggested that domestic price should be used to deflate the trade
account, which enable us to use a superlative index formula to capture the
exact contribution of real forces including terms of trade. Reinsdorf (2010)
indicated that marginal income arising from trading gains is spent in the same
way as average income, so it is appropriate to use GDFE to eliminate a trade
imbalance.
2.3 Measuring the terms of trade
There are two main approaches to measuring the Real GDP on output-side
and expenditure-side and the terms of trade. The first approach is focusing
on temporal changes in the terms of trade from the perspective of national
accounts. In the pioneer work by Diewert and Morrison (1986), production
theory and translog GDP methodology are employed to measure the changes
in the terms of trade. The measurement work can be reduced to Tornqvist
index. Based on this framework, many authors have attempted to measure the
terms of trade for one country over years (Morrison and Diewert, 1990; Kohli,
2004, 2006, 2007; Diewert, 2008; Reinsdorf, 2010; Diewert and Yu, 2012). The
second approach is to investigate the changes in the terms of trade across
time and space from the perspective of international comparison. The
traditional Gary-Khamis system has been modified to include differences in
the terms of trade between countries, which enable us to measure the real
GDP from both the expenditure-side and the output-side (Feenstra et al.,
2009). Penn World Table incorporates these technics to provide RGDPe and
RGDPo started from version 8.0 (Feenstra et al, 2015).
2.4 Comments on the literature
It is difficult to measure the total effect made by the international trade
and the consequent price movement on production and consumption based
only on an economy’s national accounting data.
Latest version of PWT uses two sets of indexes called CGDP and RGDP to
measure the real GDP on output-side and expenditur-side, but there are still
questions to ask: (1) The implicit PPP based on Fisher quantity index may not
be transitive. (2) We can’t decompose the Fisher quantity index into
meaningful parts to directly measure the effect of terms of trade.
Our questions are as follows: Can we measure the real GDP on the output-
side and expenditure-side across space and time in a more consistent way?
Can we measure the contribution of terms of trade to the real GDP growth
across space and time in a more consistent way? Based on production theory

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