Transcript Asian Historical National Accounts: Issues and Agenda

Asian Historical National
Accounts: Issues and Agenda
Debin Ma
Economic History Department,
London School of Economics
Nov. 2010
A presentation prepared for the Maddison
Memorial Conference
The Objectives of this presentation
• This presentation offers a (rather superficial) survey of
ongoing research on Asian Historical Accounts around
the world.
• But Asia is vast, even a superficial survey is beyond the
expertise of any individual (except the late Angus
Maddison).
• I will focus mostly on East and Southeast Asia, deferring
to other conference participants on South Asia
(Tirthankar Roy, Steve Broadberry, Alan Heston) and
West Asia (Sevket Pamuk) and others.
• But even for East and Southeast Asia, my presentation
refers to the works of other conference participants who
are more directly involved in the research.
Structure of Presentation
• Introduction of ongoing research on Asian
Historical Accounts for individual
countries.
• The Construction of Purchasing Power
Parity Estimates for different benchmark
periods.
Historical National Accounts for Japan
• I start with Japan which has the most fully
developed national accounts research in all of
East Asia, with national accounts annual series
running continuously from 1885 (possibly even
earlier) in continuous series.
• There has been new and ongoing research in
improving and refining the estimates.
• The new exciting research I am aware of are not
so much in revising the aggregate GDP
estimates as in extending to new areas of
research.
Two new studies on pre-War Japan
•
An example is the recent work done by Bassino, Fukao, Paprzycki, Settsu
and Yuan provided new on estimates of prefecture-level value added GDP
for five benchmark years from 1890 to 1940.
•
Their research show much smaller prefectural inequality in pre-War period
and hence revising the preceding studies of a large regional inequality in
pre-War Japan.
•
The other work by Saito and Settsu revised the labor force estimates in
agriculture, industry and services sector based on new micro-studies of
occupational structures.
•
The reallocation of labor forces across different sectors led to new
estimates of labor productivity differentials in these three sectors particularly
in Japan’s earlier stage of developments.
•
Both of these papers are presented in the recent Asian Historical
Economics Conference held in Beijing at http://ahes.ier.hitu.ac.jp/forthcoming.html.
•
New work extending Japanese GDP series to pre-Meiji?
Historical National Accounts on Taiwan and Korea
• Thanks to the Japanese colonial statistics, the Hitotsubashi group
has been the pioneer and leaders in historical national accounts on
pre-War Taiwan and Korea.
• The volume edited by Mizoguchi and Umemura (1988), Basic
Economic Statistics of Former Japanese Colonies 1895–1938,
represents a landmark in this research.
• Recently, a new volume by Mizoguchi (2008) provided a more
comprehensive national accounts statistics for Taiwan for 1900-2000
• For Korea, research by Kim Nak Nyeon et al connected with
Naksungdae Institute of Economic Research within Korea have
produced some excellent work on colonial national accounts. Their
major findings of much slower population growth rate than the original
Mizoguchi and Umemura estimates revises upward the per capita
GDP growth rates during colonial Korea.
• These databases and explanatory notes have already appeared as a
GGDC Research Memorandum GD-107 by Jan-Pieter Smits, Pieter
Woltjer and in spreadsheet form.
Pre-Communist China still remains the biggest
problem
• Here we really have a cluster of estimates for
the 1930s.
• There is an estimate for the late 1940s by Ou
Baoshan et al that should be given more
attention.
• The estimates for the 1910s relied heavily on
backward projection from the 1930s.
• The 1890s estimate originally by Zhang Zhongli
is probably valuable but needs to be scrutinized
more carefully.
Pre-Communist Chinese statistics: the way
forward
• The Hitotsubashi Group will produce a China volume, which we
hope to be a step forward.
• There are some regional GDP estimates:
– my 2008 JEH paper covers two provinces in Lower Yangzi in the 1930s,
new work by Li Bozhong covers two Lower Yangzi counties in the
1820s.
– Manchuria (or Northeast China) GDP data available for 1924-1941.
• The introduction of Maddison’s work into China (thanks largely to
Harry Wu’s efforts) has generated new research efforts within China.
That is very good news.
– Guan and Li have attempted long-term GDP series for Song and Ming
China.
– I am aware of another two scholars undertaking research on early 19th
century China.
– More collaboration and support from outside China will be helpful to
bring up the standard of research in this case.
Historical Accounts on Southeast Asia
• If the work headed by Jean-Pascal Bassino on colonial Vietnam
comes out, it will be a significant step forward. Let us hope that
works out.
• I assume we count on our Dutch colleagues for works on Indonesia
(Jan Luiten van Zanden, Pierre van der Eng and Daan Marks).
• New research on British Malaya:
– Raja Nazarin (2000, 2002, 2006, 20??): an expenditure
approach to Malaysian GDP for the colonial period.
– TIN HTOO NAING (2010): production approach to Malaysian
GDP (1900-1939)
– Ichiro Sugimoto: an expenditure approach for Singapore.
• New research: Thailand, Burma??
• We should discuss about South Asia, Central and West Asia.
II. Constructing Purchasing Power Parity for preWar Asia
• We all know exchange rate conversion have
problems when comparing GDP across
countries.
• But backward projection from later benchmark
GDP estimate also has its problems.
• Constructing purchasing power parity estimates
for pre-War Asia is a big challenge.
Fukao, Ma and Yuan (2007) show discrepancy between
current price PPP estimate and back-projected estimate
Fukao et al (2007): a list of current price PPP
estimate for East Asia
1930s Benchmark PPP reconstruction?
• Yuan, Fukao and Wu (2010) reconstruction production side
PPP for US, Japan, Korea and China.
• The above work complements the expenditure PPP but could
also connect with current research for US and Western
European (Herman de Jong et al).
• A 1930s benchmark is viable as data for price, consumption
and other information for East and Southeast Asia were far
more abundant and reliable.
• For Southeast Asia: earlier work by Bassino and van der Eng,
“Economic Divergence in East Asia: New Benchmark
Estimates of Levels of Wages and GDP, 1913–1970” is
another example.
• Possible extensions to 1910s and 1950s.
Other current price benchmark estimates
(usually based on much limited set of price
information)
• Allen et al (2011) and Allen (2009) for Europe
and China in the 18th century.
• Li and van Zanden (2010) for 1820s China.
• Van zanden (2003) for Dutch and Indonesia.
• And so on…
• Two concluding questions?
• What about the rest of Asia or the large
developing world?
• How do we link with the post-WWII
period?