THE

FIVE-PHASES OF ECONOMIC DEVELOPMENT

AND INSTITUTIONAL EVOLUTION IN CHINA AND JAPAN1


MASAHIKO AOKI
STANFORD UNIVERSITY

In the year 2010, the aggregate real GDP (PPP) of China, Japan, South Korea, and
Taiwan, China surpassed that of both North America and the European Union. According
to Maddisons well-cited estimate, these economies together also constituted the
largest economic zone in 1820, producing more than one-third of the worlds total
GDP.2 However, their share dropped by more than three quarters toward the middle of
the next century, which was then followed by the successive miracles of Japan, the
Asian Tigers, and now China. What accounts for such a dramatic fall from historical
heights and then the resurrection of the region as a whole? Is there anything unique
about East Asia? What implications does this experience have for future development?

To account for the basic mechanism of GDP per capita behavior over time and
across economies, in the past few decades development economists have been
examining the implications of endogenous interactions between technology and
demography. Core insights from their studies can be summarized briefly as follows. Over
a very long run of human history, new ideas developed as population size increased
(e.g., Lee 1988, Kremer 1993, Jones 1999). But in dominantly agrarian economies, the
fruits of technological progress were channeled into population growth, which did not
help per capita GDP growth because of diminishing returns to scale of agricultural
technology. This state is referred to as the Malthusian trap or equilibrium (e.g., Hansen
& Prescott 2002, Clark 2007). It does not, however, necessarily imply that this state lacks
dynamism.
As the state of new ideas passed a threshold point and constant returns to scale
technology free from the limits of land supply became profitable, the industrial
revolution set in with physical and human resources starting to be re-allocated to urban
industries (e.g., Jorgenson 1961, Galor & Weil 2000, Hansen & Prescott 2002). There
was also an increase in working-age population brought about by the decline in infant
1

This is to be presented as the Presidential Lecture at the XVIth World Congress of the International
Economic Association to be held in Beijing, July 4-8, 2001. I express sincere gratitude to Beth Cary,
Wenmeng Feng of CDRF, Beijing, and Yoko Yamamoto formerly of VCASI, Tokyo, for their excellent editing
and research assistance.

The aggregate share of China, Japan, and Korea in the world production in 1820 was 36.6 percent vis--
vis Western Europes 23.8 percent share. The share of the U.S. at that time was a mere 1.8 percent
(Maddison, 2006, various tables). East Asian share went down to 7.9 percent by 1950.

mortality and rise in immigration in the case of Western Europe offshoots. The hike in
GDP per capita growth occasioned by this demographic shift is referred to as the
demographic gift (e.g., Bloom & Williamson 1998) or as the population bonus in East
Asia. However, as continuing technological progress tends to increase the preference
for, returns to, and/or cost (to parents) of human capital investment, people are
inclined to have fewer children (e.g., Becker, Murphy & Tamura 1990, Galor & Weil
1996, 2000, Lucas 2002). This demographic transition leads to the modern growth
regime in which the increase in GDP per capita is sustained by Lucas-Romer technology,
if not at a rate comparable to the previous transitional phase. But this may not be the
End of History, as I will discuss shortly.
The transition from the Malthusian state to modern endogenous growth is
usually modeled after stylized facts drawn from advanced Western economies.
However, the theoretical innovation of the new approach is an understanding of
different levels of per capita income as successive stages in the normal process of
development rather than as different balanced growth paths conditioned by different
parameters (e.g., Galor & Weil 2000, Hansen & Prescott 2002, Galor 2011). From such a
unified perspective, then, the miracles of the East Asian economies are not really
miracles, but catching-up phenomena (e.g., Bloom & Williamson 1998, Ngai 2004). To
better understand the development process in general, we may also wish to know why
there are differences in the timing, duration, and institutional forms of successive
developmental phases across economies, say between the West and the East, or among
China, Japan, and Korea within East Asia. Moreover, what implications may be drawn
from these differences to unravel future possibilities of development?
As a way of introducing this discussion, let me begin by identifying phases of
development for China, Japan, and South Korea, relying only on the bare numbers of
GNP (PPP basis), population and its distribution over age groups and sectoral
employment. For the moment I will set aside institutional forms. Following the unified
approach, I will start with the Malthusian phase of economic development, or the M-
phase in short, in which agricultural employment is high, say more than 80 percent,3 and
per capita income is low and stationary. According to this simple criterion, there would
not be much argument in identifying the developmental stages of China in the late Qing
Dynasty, Japan in the late Tokugawa, and South Korea in the late Chosn Dynasty as
being in the M-phase.
A difficulty of phase identification arises in discerning the onset of the transition
to the post-Malthusian phase. For Japan it is conventional to regard the transition as
triggered by the Meiji Restoration. Indeed, GDP per capita grew at the compound rate
of 1.92 percent from 1870 to the pre-War peak in 1941, in comparison to 0.19 percent
during the years 1820 to 1870 according to Maddisons estimation in terms of 1990
International dollars.4 However, the pace of reduction in agricultural employment
3

Needless to say, in this stage a large proportion of farmers was also engaged in various non-agricultural
activities such as handicraft manufacturing for domestic consumption as well as for markets.

According to Ohkawa & Rosovsky (1973), Table 2-1, per capita GNP growth rates were 1.64 percent for
1917-1931 and 4.48 percent for 1931-37.

remained rather slow, keeping the employment level at fourteen million throughout the
pre-War period.5 Thus Hayashi & Prescott (2008) described their hypothesis as the
transition from Malthus to Solow was inhibited by the barrier to labor mobility in this
phase.
For China and Korea, how to characterize the pre-War period is a thorny
question. According to Maddison, Chinas per capita GDP growth between 1870 and
1936 was merely 0.09 percent, while the population growth rate was 0.52 percent, as if
typical Malthusian phenomena ensued.6 The share of agricultural employment
remained at 83.5% percent even in 1952. South Koreas GDP per capita almost doubled
in the period between 1911 and 1938, but it was under the colonial rule of Japan. It
sharply dropped, after the end of World War II; and the 1911 level of per capita income
was not regained until the end of the Korean War in 1953. Japans GDP per capita also
sharply declined after 1941, and did not recover its previous peak until 1956.
Certainly the tolls of imperial aggression and colonialism, the Great Depression,
World War II and the Korean War, and Chinas civil war and Revolution defy a
mechanistic application of the Malthusian criterion to the first half of the twentieth-
century in East Asia. Therefore, by leaving aside for a while the characterization of the
pre-War developmental phase of China and South Korea (or, alternatively, by regarding
that period as a kind of transitory phase), let us move on to see if data in the second half
of the century can suggest a clearer picture of the developmental pattern in East Asia.
Applying macro accounting to official data on China, Japan, and South Korea, I have tried
to identify successive development phases by distinct patterns of sources of per capita
GDP growth. The sources are: (1) demographic-economic change in the ratio of total
employment to total population, g(E/N); (2) structural transformation, g(S), composed
of the shift of employment share from the primary industry, referred to below as the A-
sector, to the secondary and tertiary industries, referred to below as the I-sector, and
relative increase of output per worker in the A-sector vis--vis that in the I-sector; and
(3) changes in per worker output in the I-sector, g(YI /I) (this last item may be further
decomposed in changes in TFP (Total Factor Productivity) and capital-output ratio,

5

The share of agricultural employment was reduced from 64 percent in 1885 to 42 percent in 1940, but it
jumped up to 59 percent in 1950 as many soldiers and civilians who returned from abroad after WWII
went back to rural areas because of the shortage of food and urban jobs.

I note that some recent studies assert that the degree of poor industrial development as these macro
figures would suggest is might be somewhat misleading. For example Eastman notes that per capita cloth
consumption nearly doubled between the 1870s and the late 1920s (Eastman 1988: p.95). Rawsky
estimates that industrial output grew by an average of 8.1 percent during the years between of 1912 and
-1936 (Rawski, 1989: pp.70-71). A previous study by Chang (1969) also provides a similar estimate of
8.439 percent growth in industrial value-added (including Manchuria) between the period 1912 and 1942.
It is to be noted, however, that industry yet occupied a small place during the four decades of the
Republican period, as the base of growth was very low, and that the linkage between the modern industry
and the rural economy remained rather tangential (e.g., Feuerwerker 1995, pp.101-121). According to
Perkins estimate, the share of modern industrial output in GDP remained at 7.463% in 1933, while pre-
modern manufacturings share was 12.4% (Perkins 1975, p.117) .


TABLE: SOURCES OF PER CAPITA INCOME GROWTH: CHINA, JAPAN AND SOUTH KOREA

CHINA

JAPAN

1870-1938
1870-1951
1952-1967
1967-1977
1977-1989
1990-1999
1999-2008

1880-1944
1880-1955
1955-1959
1959-1969
1969-1979
1979-1989
1989-1999
1999-2008

Starting Y/N
(Maddison)

g(Y/N)
(Maddison)

g(Y/N)
(Official)

G(L/N)

G(S)

G(YI/LI)

Phase

530
530
537
712
895
1,858
3,259

0.09
-0.24
1.90
2.31
6.13
6.44
-

3.53
4.26
8.12
9.49
9.32

0.76
0.28
1.44
0.03
0.30

-0.58
1.65
3.47
1.07
1.60

0.77
0.28
3.21
8.39
7.41

863
863
2,771
3,554
8,874
13,163
17,942
20,641

2.03
1.57
6.42
9.58
4.02
3.15
1.41
-

6.32
8.13
3.80
3.81
0.91
1.70

1.43
0.91
-0.41
0.23
0.10
-0.34

2.34
0.98
0.62
0.40
0.28
0.10

2.54
6.24
3.59
3.18
0.53
1.93

G
K
K/H

G

K
H
PD?

SOUTH KOREA

1911-1944
777
1.64




1911-1963
777
0.82




1963-1970
1,186
7.39




1970-1979
1,954
9.14
7.81
2.22
2.29
3.29
1980-1989
4,144
6.91
8.62
1.60
2.27
4.74
1989-1999
8,027
5.12
5.47
0.51
0.11
4.86
1999-2008
13,222
-
4.60
1.22
0.11
3.28


provided that reliable sectoral capital stock data are available).7 The above table
summarizes the results, with Maddisons estimate of the per capita GDP growth rates
for a comparative reference.8

G/K
H

The decomposition is calculated as follows. Let Y = GDP, N = population size, E = total employment, Yi =
output of the i-th sector, i= A (primary), I (second & tertiary), Ei = employment in the i-th sector, i = A, I.
As Y = YA + YI, E = EA + EI,
y = Y/N = E/N[EA/E x Y A/EA + EI/E x Y I/EI]= E/N x YI /I[ 1 ]
where = / and = [EI EA]/ EI. Let [1 ] = S, which measures impacts of structural change due to

 Together with the previous observation as regards prewar Japan, it is suggested that
the post-Malthusian stage can be decomposed into two sub-phases: That is, the first
phase of national industrialization characterized by moderate per capita GDP growth
with a moderate degree of structural transformation: 1952-1977 for China and 1880-
1956 for Japan, followed by the second phase of very high per capita GDP growth under
rapid structural transformation combined with demographic gift: 1977-1989 for China
1955-1969 for Japan. The first sub phase corresponds to the era known for the
conspicuous government involvement in industrial accumulation. So let us refer to it as
the G-phase.9

In the second sub-phase, demographic factors, i.e., an increase in the labor force
share in the total population and the shift of the employment share from the A-sector to
the I-sector contributed to between one-quarter to one-half of the very high per capita
income growth in.10 A classical paper by Simon Kuznets (1957) characterizes the
reduction in agricultural share of employment. If the employment share of A-sector goes down and/or
productivity differential between the MS-sector and A-sector is narrowed, this measure tends to go up,
having positive effect on GDP per capita y. Denoting the rates of growth of the various variables by g(.), it
holds that :
g(y) = [g(E) g(N)] + g(YI /I) + g(S)
If KMS = input of capital service in the MS-sector and MS = capital share in the I-sector is available, then
the growth of labor productivity in the I-sector can be further decomposed as
g(YI /I) = [1/(1-I )] g(TFPI) + [I/(1-I)]g(KI/ YI)
8

Maddisons estimate of Chinas per capita GDP growth in terms of 1990 International Geary-Khamas
Dollars (I$) tends to be lower than estimates based on official statistics. Since officials of provincial
governments in China are rewarded for superior growth performance (e.g., Li & Zhou 2005), they tend to
overstate growth output. Many research efforts have been made to correct this problem and they are
neatly surveyed in Cao et al (2009), together with their own results. See Young (2003) for careful checking
and adjustments of Chinese official data in general. In the calculation of Chinas per capita GDP growth
rates in the Table, the year 1989-90 is not taken into account, because there was a substantial revision in
the official estimate of employment, resulting in a discrepancy as large as 72 million between old and new
series. Likewise, Korean official data of sectoral output are available on current factor costs basis between
1970 and 1979 and then on current price basis, thereafter. Therefore, growth rates between 1979-1980
are not taken into account.

In China, agriculture provided RMB 600 billion for industrialization between 1951 and 1978, while state
investment in agriculture was RMB 176 billion (Wu 2004/2005: p.117). However, the high contribution of
per worker output in the I-sector in the early phase (the early 1950s) may be largely attributable to an
improvement in the management of industrial facilities and human resources inherited from the old
regime (e.g., Perkins 1975; Feuerwerker, 1995, pp.100-121). For Japan, Teranishi (1982) showed that the
role of financial markets in financing industrial growth was not important in the G-phase, but that of fiscal
mechanism was significant in the form of de facto subsidies to non-agricultural sector calculated as
industrial differential in tax burden -- before the WWI, and in the form of formal subsidies to non-
agricultural sector after 1923. The ratio of non-agricultural subsidies to total tax revenues amounted to
31.8 percent in the years between 1928 and 1932.

10

My accounting method may underestimate the impact of demographic gifts on GDP per capita growth,
because it measures only the direct effect of labor inputs. However, the relative increase in the working-
age population may contribute to an increase in savings as well, which can increase the capital-labor ratio

reduction in the agricultural employment-share across economies and over time as

quantitative aspects of the economic growth. In East Asia, this shift was compressed
into much shorter periods than in Western Europe,11 supplemented by demographic gift
due to the rising fertility and the declining infant mortality in the preceding G-phase.12
Thus I refer to this second sub-phase of the post-Malthusian stage as the K-phase,
reminiscent of the Kuznets process. We see that in South Korea the K-phase was
coalesced into the G-phase. This was because the mobility of the rural population had
already begun at a significant rate prior to national industrialization during the colonial
period and the years of Korean War.13

As the K-phase works out its course, the possibility of sustained GDP per capita
growth hinges on the ability of the economy to steadily improve on per worker output
in the I-sector, particularly in terms of TFP and human capital investment. Let us refer to
this phase as the H-phase, reminiscent of human-capital based, endogenous growth.
Japan appears to have succeeded in this transition in the period between 1970s and
1980s, but failed to sustain the continued growth of per worker output in the next
decade, with a modicum of turnaround in the 2000s. Surely there involved an element
of the failure of macro economic management, but I suggest below that the failure of
the institutional arrangements to respond to the emergent demographic transition has
begun to cast shadow on per capita income growth. On the other hand, South Korea has
been succeeding in sustaining high per worker output in the I-sector on the last two
decades (1989-2008). A conspicuous increase in the labor participation ratio in the
2000s is partially due to the demographic gifts bestowed by the second generation of

that enhances industrial output per labor. Higgins and Williamson (1996, 19970) estimate that the 13.6 %
upward swing in the savings rate in East Asia between 1970 and 1992 can be almost entirely accounted
for by falling dependency rate in East Asia, which raised accumulation rates by 3.4 percent and
augmented the growth in GDP per capita by 1.5 %.
11

According to Kuznets (1957), it took 84 years for France to reduce agricultural employment share from
52 percent in 1866 to 33 percent in 1950.

12

In Japan the crude birth rate stayed at high level of more than 30 percent between 1900-1947. Then it
steadily went down to less than 10 percent after 1990. In China the crude birth rate shot up to more than
40 percent in 1963 in reaction to a decrease of population due to the Great Leap Forward and remained
at the level of more than 30 percent until 1970. In the 1970s the rate steadily declined below 20 percent
even before the official introduction of the one child policy.

13

During the colonial period, the coherence of the traditional village was weakened because of the
Japanese interests in landownership and the relative decline in the price of rice, while wage-earning
opportunities outside Korea became relatively better. According to an authoritative study by Kwon
(1977), by the year 1940, 14 percent of the ethnic Koreans were living in Japan and Manchuria. A large
proportion of them (about three quarters of those working in Japan) returned to Korea after WWII and
tended to settle in the urban areas. The net rural-urban migration during 1949-1955 is estimated by T.H.
Kwon to range between 650, 000 to 750, 000 out of the total population of 20 millions in 1950. In
Kangwn Province, 21 percent of the population migrated from the rural sector. During the period 1963
1969, for which official employment data are available, the share of the A-sector employment in total
employment was already reduced from 62.9 percent to 50.2 percent. However, for this period, the
official sectoral output data on market price basis are not available (for me).

post-war baby boomers, but also reflects the rapid decline of dependency ratio due to
fertility decline: the typical H-phase phenomenon at an extraordinary acceleration.14

For the period from 1990 to the present, Chinas continuing high growth of GDP
per capita is no longer supported by demographic gifts as a result of the one-child
policy, but the contribution of the structural transformation still accounted for close to
one-quarter of per capita GDP growth in the 2000s. Whether or not the contribution of
the structural transformation will continue to persist constitutes the crux of economic-
demographic debates in China now.15 For both Japan and South Korea, the turning
points from the K-phase to the H-phase (around 1970 and 1990 respectively) are
marked by the reduction in the share of agricultural employment to below 20 percent.
In China, the share of agricultural employment in 2009 was 26.4 percent in the coastal
provinces and 46.3 percent in the inland provinces. If the 20 percent share is used as a
rule of thumb, the transition out of the K-phase may soon occur, or even have already
occurred in the coastal area, but may not be so imminent in the inland area.
Incidentally, GDP per capita (PPP basis) in the coastal provinces in 2009 is US$ 10,616,
which happens to be almost equal to the World Bank estimate of the world average,
while that of the inland provinces is 55 percent smaller (US$ 4,755). 16

China appears to continue to enjoy robust improvements in output per worker in
the I-sector. However, the figure reported in Table may be somewhat overrepresented
in this respect, as the official data may underestimate the level of labor inputs in the I-
sector as discussed by Cai and Wang (2007). (To the degree that this is the case, the
contribution of structural transformation might be even higher than the Table indicates
because of the relatively higher improvement in per-worker output in the A-sector).
How much of the growth in industrial output per worker is due to TFP/human-capital
investment cannot be known for sure without reliable capital stock data that are still
14

Between 1985 and 2005, the proportion of those under the age of 15 in total population was reduced
by more than 10 percent.

15

The debate is often phrased as whether or not the Chinese economy is facing the Lewisian turning
point. But this way of formulating the issue appears to be misleading, because the Lewisian model
mechanically combines two distinct models: the classical model of unlimited labor supply and the
neoclassical model of the competitive labor market the transition. This is not in accord with the unified
approach of recent vintage as briefly described at beginning of this paper as it ignores aspects of rational
choice by the farmers under institutional constraints that may appear to be the presence of surplus labor.
See Jorgenson (1967) for an earlier critique of the Lewisan theory in favor of a rational choice model.

16

Using official Chinese data, the division between the coastal provinces (Beijing, Tianjin, Liaoning,
Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong) and inland provinces (Hebei, Shanxi, Jilin,
Helongjiang, Anhui, Jiangxi, Henan, Hubei, Hunan, Guangxi, Hainan, Chongqing, Sichuan, Guizhou, Yunnan,
Xizang, Shaanxi, Gansu, Qinghai, Ningxia, Xinjiang) is made on the basis of a cluster analysis of correlates
between gross provincial product per capita and the share of agricultural employment across provinces.
This analysis detects only one conspicuous outlier in Inner Mongolia where the agricultural employment
share is relatively high (48.8 percent), but per capita gross product is comparable to those of coastal
provinces because of high mining output. This province is excluded in the calculation. The conversion of
per capita outputs to US$ is based on the PPP conversion ratio of the World Bank. The World Bank
estimate of the world average is US$ 10,691.

unavailable in the public domain. However, notable scholarly researches estimating TFP
have been published, which are neatly surveyed by Cao et al (2009). According to their
own industry-based study, for the 1982-2000 period TFP growth in the I-sector was 1.8
percent (2.1 percent in secondary industry and - 0.3 percent in tertiary industry) and
the major source of per worker output growth was capital accumulation rather than
TFP.

Looking further ahead, the three East Asian economies will face a significant
degree of shrinkage in the economically active segments of the population. In Japan,
where the highest life expectancy in the world has been achieved, the share of the
potential labor force (aged 15 65) is projected to decline to as low as a half of the
population by mid century. The share in China already reached its peak in 2010, and
even the absolute size of the potential labor force is projected to start shrinking after
ten years. In South Korea the share of the potential labor force will not reach its peak
(73percent) until 2015, but the pace of aging will become even faster than in Japan
thereafter.17
These dramatic demographic changes in East Asia are the consequence of the
lower fertility that is characteristic of the H-phase, combined with the extension of life
expectancy due to improved healthcare and life comforts made available in the same
phase. Furthermore, the increasing rate of human capital investment through longer
schooling tends to diminish the actual work participation by the younger cohorts. These
three trends are taking place at a greater unprecedented speed in East Asia than
elsewhere, because the transition from the G-phase to the H-phase was compressed
into a much shorter period of time. Per capita income may continue to grow, provided
that there will be a further increase in per worker output, as well as the development of
balancing forces to increase the rate of labor participation and reverse the decline in
fertility and so on. But if such a development of a technological and demographic nature
calls for substantially new ways of playing societal games, it would be apt to consider
the possibility of a new phase in economic development: the phase of post-
demographic transition, or the PD-phase in short. It is arguable to what extent the
tendency toward population aging and low fertility is universal. However, as indicated
by the fact that the sustainability of the social entitlement system and, accordingly, that
of public finance, becoming a common serious issue across all the developed
economies, East Asian countries may be just getting ahead in a new demographic
transition (Eggleston & Fuchs, 2011).

I have thus far identified five successive phases of the development process -- M,
G, K, H plus PD, which suggest a common development pattern across the East Asian
economies as well as largely consistent with the unified approach to development.
However, in spite of general commonality as regards the quantitative nature of
developmental phases, there are also differences in timing and duration of each phase
across the East Asian economies and beyond. Why? And what implications are there for
17

The South Korean share of the age group between 15 and 65 is expected to reach 73 percent in 2015. It
will then start to decline to about 55 percent over the next 35 years, as opposed to the 50 years needed
for Japan to experience the same magnitude of shrinkage.

future development? These questions evidently call for an explicit consideration of

institutions that I have abstracted so far. However, the notion of institutions has not
been easily agreed upon among economists and beyond. One of the recent approaches
popular among economists is to measure the quality of institutions of each economy by
their distance from presumably ideal institutional arrangements composed of, say, the
rule of law, generalized trust relationships, protection of minority shareholder in
corporate governance and the like, and regress economic performance on these indices.
But what does such a distance imply? Can, and ought, such distances be narrowed
simply by enlightened government policy and innovative entrepreneurial behavior so
that all the economies converge on the modern growth regime supported by those
ideal institutional arrangements? Although such diagnoses and prescriptions appear to
be clear, they may not be very helpful in understanding the nature of the paths that the
East Asian economies have been taking and in making predictions for their future
trajectories and prescribing policy for them.

Instead of exogenous view of institutions, I adopt the following conceptual
framework for understanding institutions and their dynamics, as elaborated on in my
recent works (Aoki 2001, 2010, 2011). Institutions are commonly cognized, salient
patterns by which societal games are recursively played and expected to be played. Such
patterns may be summarily and publicly represented by laws, norms, organizations,
social rules, and other external artifacts, which may be referred to as substantive forms
of institutions. The essential function of these public representations is to mediate
between the state of play and agents individual beliefs in a recursive manner. Individual
beliefs in conjunction with individuals incentives generate states of play, while the
salient features of the recursive states of play provide reasons to believe those public
representations. Through such a stable mediation of institutions between peoples
cognition (beliefs) and actual play of societal games, a specific pattern of per capita
income and demographic behavior is generated in society. As such, institutions could be
subjected to game-theoretic equilibrium analysis.

But institutions also change. Then, the basic aspects of the institutional
trajectory over the development process may be characterized as punctuated equilibria
as a first approximation, i.e., as a sequence of successive equilibria. However, these
successive equilibria are not disjointed from each other, but may be linked in a path-
dependent manner. In each phase the state of play is not in a precisely stationary
position. It is in constant motions induced by changes in the per capita income level,
demographic factors such as the age composition of the population determined by
previous generations and so on. These emergent changes generate experimental and
new ways of play in response, which anticipate and constrain transitions to the next
phase of the state of play. Then, salient patterns of the new state of play will be
summarized and publicized in the new substantive form of institutions. Thus,
institutions should be viewed as co-evolving with economic-demographic dynamics
rather than determining economic and demographic performance in an irreversible way.

This much is simple enough. However, in order to understand varied patterns of co-
evolution across economies, historical sources of variations need to be identified.18

New-Institutional Economics submits that the establishment of the modern
economy was made possible on the basis of the third party-enforcement of contracts
and property rights to which governments themselves are subjected. The transition to
this state from the pre-modern state was intermediated by the emergence of
anonymous exchanges within cities and across cities under various experimental
arrangements of contract enforcement (e.g., Greif 2006). Also, the emergence of
impersonal perpetuated organizations corporations -- is considered to provide as the
doorstep condition for the transition (North et al 2009). In contrast, it is generally
reckoned that in East Asia exchanges were traditionally less anonymous and supported
by more tangible, specific relationships, such as kinship, dynastic hierarchies, and trust
building among restricted groups, which deterred the autonomous development of
competitive market exchange. The historian Landes (2006) also notes that Chinese
technology stopped progressing to initiate its own Industrial Revolution because of the
lack of a free market and institutionalized property rights and totalitarian control
over all the activities of social life. But, he added that agriculture being the chief
exception.(p.6)

The M-phases of the East Asian countries were dominated by agrarian
economies where individual peasant families cultivated small plots, not more than a few
acres, which they owned or leased through contracts. This form of selfmanaged
peasant farming was clearly distinct from the institutional forms of pre-modern
agriculture in other economies, such as the large-scale plantations in Western Offshoots
and colonial economies, serfdom in medieval Europe (and in East Asia in the period
preceding the M-phase as defined above), and mobile cultivation in sub-Saharan
Africa.19 As discussed forcefully by Oshima (1987), Hayami & Otsuka (1993) and others,
self-management by peasants without hierarchical monitoring better fit the conditions
of monsoon agriculture that required attentive human care to vegetation in response to
the changing climatic conditions20 On a similar ecological-agricultural basis, however,
18

This procedure may be thought of as being related to the instrumental method used to quantitatively
measure the impact of institutions on economic performance (e.g., Hall & Jones 1999; Acemoglu, Johnson
& Robinson 2001). In order to cope with a possible endogeneity problem, these authors seek exogenous
sources of institutional variations (instrumental variables) in language (as a parameter measuring the
facility of access to the Western institutions) and disease environment (as a parameter affecting the
facility of settlement as colonial policy), respectively. My concern is rather to understand qualitatively the
mechanism of co-evolution of demographic-economic performance and institutions. I emphasize below
agriculture-related climate conditions and geo-political situations preceding the M-phase as important
sources of variation between institutional trajectories of China and Japan.

19

Arguably there was a modicum of similarity between the yeomen prior to the enclosure and the East
Asian peasants in terms of land-holding, self-managed farming See Pomerantz (2000) for other examples
of similarity between Western Europe and East Asia prior to the spread of colonization by the European.

20

Environmental conditions more closely match the wet farming more in the Yangzi River region of China
and Japan islands, but not necessarily everywhere in East Asia. In northern China climate conditions are
less favorable for wet farming and water transportation, while natural disasters leading to famine were

10

diverse institutional forms evolved even within East Asia with regard to the enforcement
of property rights in farmland and lease contracts, trust relations and inheritance
practices among peasants, as well as tax relationships among peasants, landlords, and
governments as an essential element of the political state. In my view, these
institutionalized arrangements cannot be simply summarized for the East Asian
economies as a whole in terms of such generic notions as Confucianism, totalitarianism,
landlord exploitation, kinships, and so on. And those differences in the M-phase, some
more obvious and others subtler, appear to be non-negligible sources for bifurcation of
institutional trajectories of the two economies and thus for their economic performance.
Let me illustrate this point as regards a couple of contrasting institutional forms
between China and Japan.

Though founded on similar family-based peasant economies in the M-phase, the
inheritance practices were different between China and Japan. In China there was the
longstanding tradition of partible inheritance among sons.21 It may be that this practice
was sustained because of its consistency with the interests of the successive dynasties
to restrain the emergence of counter-powers based on large landownership. In any
case, this practice limited the unit size of farmland ownership and made its turnover
fluid. However, even if families were forced to sell their own farmlands, they often
continued to cultivate the plots under lease contracts from the buyers. Land became the
object of investments by gentries, merchants, and better-off farmers. Thus, intricate
networks of leasing contracts evolved within and across villages and these contracts
were sold and bought like stocks even without the knowledge of the peasants
cultivating the transacted plots. However, tax obligations were placed on the owners,

more severe. Perkins (1969) argues that these conditions explain the higher rate of tenancy in the
northern China (pp. 87-98). Huang (1985) argues that dry farming produced fewer surpluses in northern
China so that tenancy rates were lower. Instead, the managerial farming in which the principal cultivators
worked together with a few laborers was more frequently observable, side by side with family farming. In
Korea an influential book on farming instructions, edited under the direction of King Sejong (1418-50),
advised against the use of wet farming because it was risky and vulnerable to rain shortages. However,
this problem was overcome by the development of irrigation systems in the late eighteenth and
nineteenth centuries, which was accompanied by an important institutional transformation of the
traditional serfdom regime under the rule of the yangban. Yangban was mandarin-like gentry with
bureaucratic backgrounds in the dynasty and they owned nobi (serfs) as the objects for inheritance and
sale (an authoritative study of the late Chonson Dynasty by Palais 1996 thus translates nobi as the slaves).
This regime gradually evolved into a family-based regime of peasant cultivation through the upward
mobility of commoners and nobi to higher social status. It was made possible by the discovery of family
lineage records by commoners and even nobi, on the basis of which they purchased free status from the
government. By the mid-nineteenth century the proportion of local yangban is estimated to have
become close to a majority of the local population. See Aoki (2001) pp. 55-58 for a comparison of Japan
and Korea in terms of history vs. ecology in institutional evolution in the M-phase.
21

The only customary exception to the equal division among sons was to set up lineage trusts, especially
in the form of land ownership (e.g., Ruskola, 2000; Zheng, 2001). This practice obviously was for avoiding
the minute division of kinship wealth and for preserving it in perpetuity. This exception was legitimized as
fulfilling the responsibility of ancestral rites that was consistent with the orthodox legal tradition.
However, most farming households were too poor to place land as a perpetual trust.

11

some small and some large. How, then, were rental contracts and tax obligations
enforced?

In this regard, organizations called the landlord bursaries (zuzhang) active in the
late Qing and early Republican periods in the advanced Jiangnan region are illustrative.
The workings of these organizations were documented and analyzed in a book of some
700 pages by Muramatsu (1970) who examined numerous private land-lease contracts
and their enforcement records housed at the Harvard-Yenching Institute, Japans
National Diet Library, Toyo Bunko Library and so on. As his study appears to be relatively
unknown to non-Japanese readers, I take it up here as illustrative of a prevailing feature
of contract-enforcement mechanism in Chinas M-phase.22 These organizations acted as
agents for multiple landlords who owned large numbers of small plots of land widely
scattered and mutually intermeshed. They collected rents from hundreds, sometimes
thousands, of peasant tenants, paid taxes to magistrates and received fees for these
services.23 They were normally created by, and served, gentry families, but also
entrusted by other landowners even of different clans. Thus, although family metaphors
and ancestral rites were often invoked to perpetuate their activities beyond a single
generation and to be politically correct, they may be considered to be a quintessential
example of corporations a la Chinese style, or what Ruskola (2000) calls a clan
corporation.24 Namely, landlord participation in them was voluntary rather than
natural kin-groups based; members drew the benefits of a steady stream of rents from
their activities that otherwise would be costly to secure; they were perpetual beyond
the lifespan of any natural person; and they internalized administrative structures
independent of particular persons. They were even equipped with a small army of the
physically strong to literally enforce rent payments, while relying on the legal/physical
assistances of the magistrates, whenever there was a need to punish rent arrears and
settle contract disputes on their own behalf.

22

A brief English summary of his research is found in Muramatsu (1966), which pointed out that the same
institutions had been studied in Fei (1939/1946) under the name of a rent-collecting bureau, which I
have not yet had a chance to read. There are brief references to the Muramatsus article in Perkins
(1969), p.102, Feuerwerker (1980, 1983/1995), p.25-6, 35, and Zelin (1986). Also see Eastman (1988),
p.78n.

23

According to Muramatsu (1970), about 20 30 percent of rent revenues were paid as taxes, 10 percent
to the bursary as a fee, and the remaining 60-70 percent were to landlords in the late nineteenth century.
However, the share of landlords started to decline dramatically after 1920, while tax shares went up
(pp.31-43). For this, see also Perkins (1975), p.124.

24

Although Ruskola basically applies this concept to the lineage (ancestral) trusts, his elaborate legal
analysis of their nature can be even more forcefully applied to the landlord bursary. He describes this
notion as follows: clan corporations vehement insistence on kinship as its organizing principle did not
mean that they were just family affairs. Rather, kinship was often a finely wrought legal fiction that
legitimized the existence of private enterprises by profit-seeking individuals in a state in which
Confucianism was the official orthodoxy(2000:1617-8). He provides evidence that in the case of ancestral
fund ownership interests were even transferable. Also see Zelin (2009) and Zelin et al (2004) for the
critical role of contracts and property rights, not necessarily kinship-based, in Chinese development.

12


There were thus strategic complementarities between dynastic administration
and the landlord bursaries. For the Qing dynasty of nomad origin that had only a weak
power basis in the rural areas,25 endorsing/assisting the (coercive) enforcement of
private contracts by the latter was a way to secure tax collection. On the other hand, the
large landholders were able to legitimize the forceful collection of rents by acting as
quasi-public agents to collect for the dynasty.26 Although I draw on this case from the
rural economy, essentially the same practices appeared to have prevailed in the domain
of commerce as well. For example, they may be considered to correspond to what some
Sinologists call the brokerage concept of administration: the governments use of
local-elite leadership as brokers to solve regulatory problems with limited government
resources.27 In order to pursue this line of argument below, let me use the world
corporate body in reference to any organization in perpetuity either in political or
business.28 In recognizing the interpenetration of the state and intermediate corporate
bodies of property owners as an important element of M-phase institutional
arrangement, one question arises. As I noted, there were extensive contractual
relationships among small peasant families as well who were not able to rely upon such
quasi-public mechanisms. Then, how were private contracts among those enforced? An
answer to this question may be more sharply highlighted in a comparative perspective.
With this in mind, let me now turn to the contemporaneous scene of Tokugawa Japan.

The administrative structure of Tokugawa Japan was composed of about three
hundred, semi-autonomous Domain (Han) governments with the Tokugawa-Shogunate
(Bakufu) at the apex of the structure (Baku-Han regime). In my view, the nature of this
construct can be characterized more as a quasi-hierarchy or as a quasi-coalition than as
that of a rigidly centralized hierarchy, although there was no exit option for any domain
government from the structure. There are two aspects to this. First, the power of the
Bakufu vis-a-vis the domain was based only on the threat of terminating the jurisdiction
of any Han government in the case of serious judicial offense to this political order. Such
penal actions were actually exercised in only a few minor cases. On the other hand, each
Han had exclusive rights of collecting a fixed amount of the tax set in terms of quantity
of rice from each village under its jurisdiction. Otherwise, the Shogunate was not to
intervene in the internal affairs and rules of domains, while in turn the domains were
not to intervene in the internal affairs of villages, as long as villages collective tax
obligations (mura-uke: village contracts) were met. Landownership by farmers was
registered with the self-governing village office (mura-yakuba), and transactions on
25

According to Hsiao (1960: 50), there was only one district magistrate per 250,000 people.

26

In the northern China the magistrate and the village were in more direct contract (by the reason as
suggested in note 18), which had an important implication for the transition to the Phase-G.

27

See, for example, Duara (1988: Ch.2), Eastman (1988; Ch. 6).

28

Aoki (2011) provides the following generic definition of corporations: Corporations are voluntary,
permanent associations of natural persons engaged in some purposeful associative activities, having
unique identity, and embodied in rule-based, self-governing organizations(p. 4). It discusses varied
substantive forms incorporating this generic property that co-evolve with specific social and political
institutions.

13

farmlands were in principle possible only within the village, and not beyond the border
of the village.29 Property rights disputes within the village were legally appealable to the
magistrates office (daikansho), but in practice they were resolved within the village
thorough the mediation by the influential household heads who served as village
officials.30 The separation of samurai power from the village and the quasi-coalitional
governance structure may be considered as a stable geopolitical solution to the century-
long Warrior period (1493-1615) in which fierce competition for political hegemony had
been waged among rural-based samurai powers.

By being in the position of residual claimants after the payment of the collective
tax obligations, village members had common interests in building and sustaining
farming infrastructure, such as irrigation networks for wet farming and mutual help in
farming activities, etc. To control freeriding over collective efforts, a strict social norm of
compliance in cooperative actions was imposed on member households with the threat
of social ostracism for deviants.31 This institutional arrangement was facilitated and
made effectuated by making the households the basic unit of economic and political life
in the village, and primogeniture became the inheritance norm.32 This practice
encouraged the outflow of non-heir sons to neighboring cities, which contributed to the
growth of world-class cities in the eighteenth century.

The difference between China and Japan in terms of state-peasant relationship
had parallel impacts on the mode of trust relations among peasants. The kind of norm
that evolved in the context of the village in Tokugawa Japan, as noted above, may be
characterized as categorical, in that mutual obligations and trust were directed towards
all the members in the village and only towards them. Membership in the village
defined the necessary and sufficient conditions for the applicability of a norm of
cooperation and mutual monitoring. Categorical norms and quasi-hierarchical
governance may be then considered to be institutional complements.33 The village tax-
contracting system promoted the incentives for member farmers as residual claimants
on the one hand, while peer monitoring among member farmers within the village
made tax collection secured for the governments on the other hand.

In contrast, in the Chinese villages where the turnover of ownership of farmlands
was frequent and mobility across socio-economic strata was fluid, collective interests
encompassing all members of the village were comparatively more difficult to evolve.
Yet, there were certainly needs for horizontal economic relations among peasants in
terms of mutual help in farming and household affairs, money-lending in times of need,
29

Toward the end of Tokugawa period, some farm lands were placed as collateral for farmers loans from
urban merchants and in the case of contract default de facto ownership was transferred to the latter in
spite of repeated ordinance by the Bakufu government to prohibit the practice.

30

Ishii (1966), Watanabe & Gomi (2002)

31

For this see Aoki (2001), Chapter 2.2.

32

The Bakufu issued several ordinances restricting the division of small-sized farmland by individual
farmers in the late seventeenth century and thereafter.

33

For the concept of institutional complementarities, see Aoki (2001), pp.225-229, 267-274.

14

and even land-leasing in response to idiosyncratic family circumstances. Reciprocal

relationships accommodating these exchanges had to be strategically supported by
mutual investments in social capital among individuals having potentially common
concerns and stakes. Further, to make such specific reciprocal relationship enforceable
and credible, they needed to be monitored by third parties who themselves were a part
of the linkage of such relationships. The mode of trust relations that embeds private
contracting within a specific network of people may be characterized as selectively
constitutive, in contrast to categorical. Unlike the norm of categorical trust, it needs to
be constructed pro-actively by individuals. From a comparative perspective, I posit that
such reasoning can reveal the essential nature of the so-called guanxi (social relations in
Chinese).34 Both categorical and selectively constitutive trust relationships are
conceptually contrasted with the notion of generalized trust based on the expectation
of unspecified obligations of reciprocity in a population and supported by the
internalization of norms.

Guanxi is often regarded as essentially kinship-based, downplaying the strategic
aspects of its construction through conscious mutual effort to invest in individual social
capital. However, kinship is not an assurance of automatic fulfillment of mutual
obligations and trust, although the metaphor of kinship is often mobilized to induce and
reinforce such obligations and trust. A rich fieldwork on pre-War northern villages by
the Mantetsu research group records many telling stories about the declining role of
kinship relations in contract enforcement and trust-relations in the village.35 For
34

For a similar comparative analysis of guanxi, see Herrmann-Pillath (2010). The terminology categorical
is due to him (in his case categorical collectivism). There is an on-going controversy among
anthropological Sinologists as regards whether guanxi is emotion-based or strategy-oriented (e.g., Gold et
al, 2002). However, they do not need to be considered as mutually exclusive. From an inter-disciplinary
comparative institutional perspective, trust relationship may be theoretically conceptualized more
broadly than the reputation effects in the economic transaction domain. Suppose that agents in a
particular domain of the societal game (e.g., a village, a population) exchange social symbols such as
words, gestures, gifts, help, etc., in order to impact on others emotional payoffs (ganqing in Chinese). If
one does so with the expectation of reciprocity from others, then such actions may be regarded in the
reduced form as investment in ones own individual social capital. As evolutionary biology indicates, the
human being is innately endowed with the capacity to infer others intentions and their behavioral
consequences that other species do not have. Also, as recent neuro-scientific research confirms, there are
trade-offs between emotional payoff and material/hedonistic payoff, as if neuro-currency (Montague
and Berns 2002) is being circulated in the network of neurons within the individual brain (e.g., Fehr and
Camerer 2007, Izuma et al 2008). People may then refrain from freeriding on others collective efforts or
pursuing exclusively self-interests at the expense of others in economic exchanges, if they feel they will
depreciate their own social capital in doing so. Thus social norms and trust relationships may evolve as an
equilibrium outcome of the linked games between the domains of social-exchange and other societal
exchange, embedding and regulating actions in the latte (Aoki 2010, Chapter 3 and 4).

35

Voluminous surveys of northern and east-central China villages, the Mantetsu Survey, were conducted
between 1935 and 1942 by researchers of the Northern China Economic Research Institute of the South
Manchuria Railroad Company (Mantetsu), the largest Japanese company that was instrumental in Japans
imperial rule. Mantetsu contributed a quarter of the Japanese governments tax revenues in the 1920s
(Young 1999). The teams of researchers that included ex-Marxists students and later-to-become-
respectable-scholars recorded voluminous interviews with farmers, county chiefs, tax-contractors,
policemen, and so on, without interviewers opinions. About forty years later, some of those sites were

15

example, even between a mother and a son, leasing had to be collateralized: if her son
did not have money, the mother would rather lease her plot to another, even if he was
of different lineage, and so on. Such observations clearly indicate that the culture of
contracts was strongly solidified and prevailed widely.36 What kinship could do was to
provide better information about farmers interlocutors within which contracts took
place; it also provided a sanctioning mechanism for defectors. So it is not surprising that
contract existed within the context of extended kinship. However, it did not guarantee
that transactions would occur; it simply provides a more favorable context within which
to negotiate a contract. 37

I have discussed representative institutional arrangements in the M-phase of
China and Japan in terms of state-peasant relations (taxation on farmlands) and norms
regulating agents behavior. Let me recapitulate their essences. Chinese farmers did not
have a beneficial access to the quasi-statist corporate body (such as the landlord
bursary), but they strategically substituted guanxi-embedded private contracting for it
(and for the neutral state that did not exist). On the other hand, the large property
owners were able to strategically link an access to such a body with mutual investments
in their own guanxi so that payoffs from each of them were reinforced. Thus, the
vertical and horizontal relations together weaved social fabric in intricate manners. In
Tokugawa Japan, on the other hand, the samurai bureaucrats who resided in the Baku-
Han capital cities and the peasants in the villages were separated physically, socially,
and legally from each other. As a consequence, the village contracting system
(accordingly, the quasi-coalitional political governance) and the categorical norms of
cooperation in the village evolved as complementary institutions. These differences in
the strategic nature of M-phase institutional arrangements was, as already noted, in
collective responses to irreversible historical and geopolitical conditions prevailing then
in China and Japan respectively. How would they in turn leave their traces on
subsequent institutional trajectories?

I mentioned that the initial phase of the post-Malthusian stage in East Asia is
characterized by strong government intervention in industrial capital accumulation. This
is the theme that was extensively discussed in the development literature in the past.
However, even in England where the transition is considered to be more privately

re-visited by Huang, then a Stanford researcher, and he confirmed the reliability of the survey in spite of
its undeniable intelligence-gathering aspects and described agrarian development between the two
periods in his own book (1970). He includes a detailed description and critical assessment of the Mantetsu
Survey in this book, pp.34-43.
36

For the important role of privately enforced contracts in Chinese development vis--vis kinship ties, see
Muramatsu (1949/1975), Ruskola (2000), chapters in Zelin et al (2004).

37

In passing, I speculate that kinship may have provided a much stronger basis for trust relations in Korea
than in Japan and China. Categorical norms may be comparatively harder to evolve when the members of
the village are less homogenous (cf. note 18), while the fast and early mobility of peasants family
members out of the village (cf., note 12) may make credible information sharing and monitoring relatively
less costly within a kinship group (and also possibly among people who have the same home village).

16

motivated and driven, the role of government was not negligible in her own style.38 So I
pose a question as to how the institutional differences in the M-phase between China
and Japan cast path-dependent shadows on the mode of governments role in transition
to the G-phase in guise of institutional change. The literature of economic history has
established that handicraft manufacturing and production of various cash products
developed in rural economies of Japan and China to an extent not inferior to Western
Europe as late as 1750 (e.g., Tanimoto 1989, Pomeranz 1993, 2000, Clark 2007). But
from there a factory-based production system did not evolved in a way to move the
industrial revolution in the sense referred to in the beginning. In China, when merchants
who intermediated market transactions became wealthy, they were more interested in
human capital investment to achieve the status and prestige of the scholar-official class,
spending on conspicuous and cultural consumption and investing in money-yielding
farmland and lending. In Japan, the nature of quasi-hierarchical state that separated
samurai-bureaucrats, farmers, handicrafters and merchants as distinct social strata
prevented anyone from combining talents for industrial development. Thus, when
Western technology that fit the factory system became exogenously available as a
potential threat to the independence of the state and the integrity of the society, its
actual adoption called for a transformation in the substantive form of the state.

This transformation might have been comparatively easier in Japan because of
the quasi-hierarchical, quasi-coalitional nature of the Tokugawa political regime. It took
the form of the take-over of the state by lower-class samurai-bureaucrats of four Han
(eventually two after a civil war between them) , by making the restoration of Emperors
rule as a legitimizing device. The new Meiji government abolished the Baku-Han quasi-
coalitional regime and attempted to centralize administrative functions, formerly
encapsulated within each Han government, to promote the formation of integrated
national markets. It decreed that ownership of land (including farmland) be registered
at the national registry, and any dispute over property rights and breaches of contracts
be settled by the courts according to law. In lieu of the village contracting system,
farmland taxation was fixed in monetary terms and imposed on individual landowners.
However, these seemingly market-oriented reforms gradually generated unintended
consequences under the pressure of categorical norms still prevailing at the village level.

Farmers of small landholdings were exposed to deflationary pressure in the early
1880s, which resulted in an increase of the proportion of tenancy lands from 20-30
percent in the 1880s to more than 40 percent in the 1890s.39 Those who amassed
landholdings tended to become absentee landlords like their contemporaneous
counterparts in China. However, instead of relying on a corporate body of their own, like
38

For example, the enactment of the Joint Stock Companies Act 1844 allowed incorporation to be
organized by the mere act of registration without getting a special charter, which led to the development
of corporate market economy (Ahlering & Deakin 2007, Aoki 2011). See Pomerantz (2000) for a
noteworthy comparative view on the role of visible hands in the transition from the Malthusian phase
to the industrial revolution in the West.

39

Western scholars often erroneously believe that the absentee landownership was historically dominant
since the Tokugawa period, but this was not the case.

17

the landlord bursary in the M-phase of China, they relied on the traditional categorical
norm of the villages to legitimize their rent-capturing positions. While contributing
collective goods, such as schools, festivals, scholarships for able children of farmers and
the like to the village as if they were paternalistic guardians of the village, they made
use of reputation concerns among tenant peasants for the enforcement of rent
payments. As already noted, between the Meiji Restoration and the end of the WWII,
the number of agricultural employment remained constant at 14 millions, reflecting the
continued practice of primogeniture and the social norms of mutual monitoring among
households. Hayashi and Prescott (2008) estimates the economic effects of the
restraints on the free mobility of labor out of agriculture cost 40 percent of per worker
industrial output during the period, blaming the Confucian-spirited civic law imposed by
the Meiji government (actually it emulated French civic law).

From the 1930s on, the position of absentee landlords became increasingly
encroached upon by market forces as well as by government policies to protect peasant
farmers to ensure the security of the food supply and societal stability. This tendency
culminated in the post-war Land Reform that redistributed the ownership of all
farmlands to farming households. By securing property rights in land ownership, farming
households were able to choose their working hours in agriculture and elsewhere on the
basis of economic calculations, paving the road to the K-phase. The new generation of
farm households left the rural landscape behind and joined corporate organizations in
the cities after their schooling. As well known, the mode of the categorical norm of
cooperation was transplanted into these new work environments, contributing to the
high growth of per worker output during the K-phase and the heyday of the H-phase.
Later I will touch on its eventual consequences in the face of transition to the PD-phase.

In contrast, the transition to the G-phase in China was not so smooth because of
the nature of its political state in the M-phase. The late Qing dynasty made some efforts
to adopt a kind of bureaucratic-led industrial policy, such as promotion of merchants-
managed firms under bureaucratic supervision (guandu shangban), establishment of
government-business joint enterprises in arsenals, shipyards, and so on (guanshang
heban) as well as to invitations to merchants (zhaoshang) to engage in the
transportation of the products of government monopolies such as salt.40 Some of these
became profitable when managed by able bureaucrat-business persons, but most of
them were short-lived. After the collapse of the Qing dynasty, there was a brief period
when private businesses were emergent. However, military groups, political and
imperial powers soon began to compete for their political control over the commercial
and industrial domains. By 1947 the Nationalist government came to control more than
two-fifths of total industrial production by the return to the tradition of guandu
shangban. But their control over rural areas did not develop in such a way as to mobilize
resources from there for industrial development. The Nationalist government gave up
the land tax to the Provincial governments. The provincial governments then
subcontracted land tax collection to county chiefs whose positions were occupied by ex-
40

For business management of companies under these industrial policies, see Fuerwerker (1958),
Eastman (1988: chaps. 7, 8), Zelin (2009).

18

gentry or new rural elites who had guanxi with military elites in the Provincial
governments. After fulfilling the negotiated contractual obligations to the Provincial
governments, the county chiefs were able to claim the residual of the collected taxes,
which motivated them to squeeze the share of landowners rents including those for the
small landholders by the use of police force.41 Thus, the basic structure of the
interpenetration of weak state power and private interests of strong intermediate
corporate bodies (in this case, county governments) remained in a manner analogous to
the M-phase.42

The G-phase finally came to be institutionalized in the form of the Peoples
Communes in the late 1950s posterior to a brief episode of the egalitarian land reform
after Revolution. Apart from its political and ideological connotations, economic
relationships between the government and 120 million rural households were made
direct with the Peoples Communes as the exclusive intermediary bodies.43 This was in
essence the wholesale incorporation of rural households at the expense of the
traditional culture of private contracting. In this way, economic and demographic
features of the G-phase transpired la Chinese. Material resources for industrial
accumulation were extracted from the agricultural sector through direct agricultural
taxation and state-monopoly procurements and then invested in state-owned
enterprises, another types of intermediate corporate bodies, in the form of direct
subsidies and price-controls

On the other hand, the exit of farmers from agriculture was restrained by the
mandatory membership requirements (hukou) in the commune. Their incomes were
basically determined not by their marginal products but by per worker output (with
some differentials) after tax payments to the government and various collective
investments. That might be one of the important incentive reasons for the hike in
fertility in the 1960s,44 of which unintended consequence was a demographic gift to the
next K-phase. Further, the collectivization of farming made social relations among peers
at the level of the production teams (with an average size of between twenty to thirty
households) relatively more inclusive rather than selective. This made possible all-
inclusive collective actions such as the adoption of new crop varieties and chemical
fertilizers, investment in water control, tractor plowing, and so on, which were not
possible during the previous transitory phase. Indeed, between 1970 and 1977, per
worker output in the A-sector increased by a compounded annual rate of 2.32 percent,

4141

The Mantetsu Survey on land tax collection at the county level is extensively cited in Muramatsu
(1949/1975: 137-45), according to which the expenditure for the police force in a representative county
of Shandong province amounted to one-third of land tax revenue in 1941. Huang (1985: p277) also
reports that expenditures for police and military guards in one county of Hebei province exceeded 60
percent of its official budgets.

42

This statement excludes any consideration of communist-controlled China during the period.

43

See J. Lin (1994). For the political aspects of Peoples Communes, see Wu (2005), Chapter 3.

44

Total fertility rate rose as high as 7.5 percent in 1963.

19

in comparison to stagnant a 0.21 percent in the I-sector.45 Thus a stage for the transition
to the K-phase was set. The actual onset of the transition was triggered by the
restoration of private contracting, not by political design: that is, experiments at the
village level to contract out the collective obligations of tax payments to member
households through the subletting of village-owned plots to them. Indeed, to quote
Professor Jinglian Wu, Chinese reform started from the village, from the bottom, and
we may even say that it was the invention of farmers themselves. (Wu. 2007: p. v)46.

I have already referred to the quantitative achievements after the transition to
the K-phase. While impressive performance in terms of GDP per capita growth has been
making since then, various institutional issues have been presented and many of them
have been responded in ways that facilitate growth. Presently, institutional reforms in
the areas of inequality, real estate, labor shortage, local finance, and so on are
becoming widely and earnestly discussed to make the compounded transition to the H-
phase and to the PD-phases facilitated and secured. Obviously, I am not qualified to add
anything substantive to the debate among Chinese economists. Let me briefly touch on,
however, how the framework that I have been presenting might be relevant. Specifically,
I wonder if aspects of those issues are interrelated in a manner inherent to the crucial
question of relationships between the state, intermediate corporate bodies, and people
in a new guise. As representative intermediate corporate bodies, we may think of local
governments and state-owned corporations, but I will focus on the former.

In the last twenty years or so, massive migration from the A-sector to the I
sector has occurred in the form of the so-called floating population amounting some
200 million people and will still be occurring. The 2007 Property Rights Law stipulates
that the farmers contractual use-rights to village-owned farmland are extendable up to
30 years, and includes the rights to lease and sell the rights. Thus, de facto ownership
rights appear to have been endowed to households with rural hukou. The important of
this can be confirmed in the light of comparative perspective, as suggested by the case
of post-WWII land reform in Japan. Yet, the households who now reside and work in the
urban areas but still have rural huko (household registration) may find the opportunity
costs of selling the subcontracting rights to be too high. First, their employment
opportunities, social security packages including those for their parents remaining in
rural areas, equal opportunities for children to proceed to higher education, and so on
are not secure enough, even though there have been notable improvements in certain
respects, especially in the big cities. Second, opportunities for farmers to realize capital
gains from the sale of the contractual use-rights may be limited in practice, because
markets may be under de facto monopsonistic control of local governments. Indeed, a
large share of fiscal revenues of local government amounting more than one-fifth of the
total are financed by the development surplus realized from their acting as local
45

The rather large contribution of structural transformation in this period reported in Table 1 reflects this
relative increase in per worker output in the A-sector rather than emigration from the A-sector to the I-
sector.

46

See Wu (2004/2005), chapter 3 for detailed historical descriptions and economic and political analysis
of the household contracting system. Also, Lin (1993) is an important contribution to this subject.

20

monopsony vis-a-vis farmers, while selling use-rights of land to urban developers

through auctions.

But there is a cause to it. Local governments have constitutional obligations to
provide social security and social protections in the areas of health, education and the
like to people with hukou under their jurisdiction. On the other hand, their fiscal basis
became seriously squeezed through the public finance reform in the 1990s to make
integrated national markets by centralizing revenues from value added taxes and
income taxes. In that sense, the origins of Chinas real estate boom may be considered
to be partially fiscal rather than purely financial. However, the responsibility of local
governments to provide welfare benefits will increase as the population ages.
Particularly, the 1960s baby-boomers who remain in the rural areas may find it
increasingly difficult to migrate because of their age and skills. Who will care for them
after ten years or more? The prospective transition to the PD-phase may make the
scope of kinship in social protection diminished.

Thus, the challenge that China is facing is compounded: it is not only an
intermediate corporate body problem inherent in tradition, nor is it only labor and
housing shortage problems emerging out of the Lewisian turning point from the K-phase
to the H-phase, nor is it only a question of how local governments can finance equitable
welfare programs to cope with the coming of the PD-phase before China becomes
affluent(Cai and Wang 2006). Indeed, all these issues are mutually interrelated, for
which solutions need to be complementary as well as path-dependently viable. They are
newly emergent problems, but at the same time uniquely traditional problems in that
the role of intermediate corporate bodies (in this case, local governments) between the
state and people is the underlying issue.47 These problems can be therefore solved
gradually by the wisdom of the Chinese people, not over night by a mere emulation of
Western style rules.48

Guanxi will continue to play an indispensable role supporting the social fabric in
the process of searching for a solution, while it will adapt its substance in linkage to
economic and demographic changes that will unfold in the process. It is hard to predict
how this traditional institution will evolve. But one thing appears to be clear. As kinship
relationships decline in its scope, the constitutive nature of guanxi as trust-relationships
47

Qian & Weingast (1995) also focuses on provincial governments as a positive driving force of growth in
the Chinas development prior to a fiscal system reform in the mid 1990s. See also Li & Zhou (2005).

48

One solution could be to allow the migrating households to enjoy the full benefits of urban citizenship
in exchange for the sale of subcontracting rights to farmland, if they so choose. Capital gains from sales of
subcontracting rights may be taxed to finance part of the fiscal obligations of local governments. Also, as
the zoning of farmland is desirable to secure food supplies and to prevent disorderly private
development, the role of local governments in regulating real estate transactions will remain
indispensable in one way or another. Then, one of the crucial questions is, as publically well-recognized,
how to make relationships efficient, fair, caring, and transparent between farmers and urban citizens, on
the one hand, and provincial and county governments on the other. For this, it appears to be crucial to
design the tax system in the areas of property rights, social security, inheritance and so on, as well as
fiscal transfer scheme between the central government and local governments, to make the latter fiscally
viable under transparent rules. Cf. Cai & Wang (2006, 2008), Tao & Shi (2010) for related proposals.

21

will further transpire. According to a demographers estimate, the size of the typical
kinship network has fallen to about 10 percent of what it was a few decades ago
(Tuljapurkar, 2010).49

Let me switch back to Japan. The challenge of how to adapt the substantive
forms of institutions to emergent situations is not simple for Japan either. The two
features I have extracted as the inherent nature of Japanese institutional arrangements,
quasi-hierarchical nature of governance and categorical norms, still permeate the ways
societal games are played in contemporary contexts. They manifested themselves most
dramatically in the recent March-11 disaster. When the natural disaster of the
earthquake and tsunami shook the North East Coast of Japans main island, four nuclear
reactors at Fukushima Daiichi Nuclear Plant of TEPCO shut down automatically in
seconds. The reactors were then supposed to be cooled down so that a critical reaction
would not be triggered by decay heat. However, soon it became clear that all power
sources to the cooling system had been knocked out. The imminent question was who
had the ultimate authority and responsibility to decide what to do at this critical
moment; but this was not clear to anybody. Between the Prime Ministers Office, safety
regulators, TEPCOs top management, and the plant manager, there were continual
exchanges of words, mutual guess-work and suspicions about others intentions,
hesitations on taking action, unwillingness to disclose unfavorable information, and so
on. Very soon, meltdowns in the reactors and subsequent hydrogen explosions occurred,
as the opening of the vent valves to release pressure built up in the reactors was
delayed, perhaps because of the lack of preparation as well as by hesitation to do so.
The whole picture of the situations has yet to be made clearer. But this episode is telling.

During the Three Mile Island crisis President Jimmy Carter went to visit the site,
primarily to calm the public. Although he had been a nuclear submarine officer and had
experience with pressurized water reactors, he wasn't there to direct things. The plant
manager was given ultimate authority and finally resolved the crisis by opening the vent
valves on his own judgment. In spite of major social tensions at that time, the actual
radiation emission was kept to a manageable level. When Chernobyls Water Coolant
Reactor exploded because of mishandling by site-engineers, Gorbachev, who was at the
head of the chain of command, kept his silence for eighteen days. He eventually sent in
500,000 so-called liquidators, composed mainly of Soviet Army Reservists, virtually
unprotected to shovel off highly radioactive graphite debris. There is still no account of
their health status today.

These three episodes, albeit in extremely critical circumstances, are remindful of
three prototypes of system models that have been analyzed by economists, system
analysts, system engineers, and others. Imagine that systems are composed of multiple
modules distinct in their functions, tasks, and so on. Systems may then be distinguished
by ways in which the functions/activities of these modules are coordinated: that is,
either (1) by open interface rules designed ex ante, while the workings of each module
49

The size of a kinship network is proportional to the square of the total fertility rate (Goodman, Keyfitz,
& Pullman 1974). As Chinas TFR has fallen from 5 to around 1.5 from the G-phase to the K-phase, the
conclusion follows

22

are encapsulated from each other to make the best use of expertise internalized in it as
far as they follow the interface rules (open-rule-based modular system); (2) continuous,
on-going negotiations among agents in charge of modules on their outputs, while direct
interventions in the internal workings of others are mutually refrained from
(negotiation-based modular system); or (3) by hierarchical chains of downward
commands and upward reports (classical hierarchy).
This three-way comparison of system design, though extremely stylized, may still
entail a few important implications. It is well established that the open-rule-based
modular system is superior in its self-organizing innovative capacity, because
constituent modules can be evolutionarily substituted, added or superseded by
improved ones as far as the latters functions are consistent with the open interface
rules.50 Further, it may have advantage in dealing with large system shocks in the world
of complexity, provided that some modules encapsulate highly specialized functions
that can be triggered in response to signals indicating emergency. On the other hand,
the negotiation-based modular system may be superior in adapting its systemic output
performance to mildly changing environments. Theoretically, there are various trade-
offs in performance characteristics, like this and others, between the three proto-types
of system design (e.g., Aoki 1986, 2011, Baldwin and Clark 2000). Although any effective
organizations in the real world may be hybrids of the three to varying degrees, it is clear
that the complementary arrangements of quasi-hierarchical governance and categorical
norms would have the best affinity to the continual negotiationbased system.
The March-11 nuclear disaster revealed aspects of comparative weaknesses of
the corporate form internalized in TEPCO. They integrate the generation, transmission,
distribution, and retail of power as a regional monopoly and have a strong tendency
toward the continual negotiation-based system. The seemingly seamless coordination
among these varied functions may have performed well in terms of minimizing the risk
of power outage in mildly changing environments. But when they faced this severe
crisis, negotiated responses among concerned agents failed to contain the impacts of
the natural disaster to a moderate level. Even more importantly, the failure of power
sources for the coolant system may not have been just the consequence of this natural
disaster of a magnitude beyond human imagination. Warnings of possible disaster of
that magnitude had been expressed during the preceding years in official meetings of
the government to discuss safety regulations, debates in the Diet, as well as specialists
writings based on research in historical documents and geological engineering findings.
Yet, the entrenched group of nuclear specialists within TEPCO and their academic allies
had not heeded these warnings, while regulators as well as top management of TEPCO,
lacking expertise in nuclear engineering, did not dare intervene. Nuclear energy
specialists share a norm of cooperation categorically specific to their profession. It is
telling that this entrenched group is nicknamed the nuclear power village.

50

This property suggests the essence of comparative advantage in the area of industrial innovation of
Silicon Valley cluster of venture businesses vis--vis the monopolistically integrated old IBM, a
quintessential case of the classical hierarchy. For this, see Baldwin & Clark (2000)

23

The March-11 disaster disclosed some essential problems inherent in the

Japanese system, political and corporate, in acute ways. At the same time, however, it
may suggest a certain degree of hope in direction as well. The Japanese people were
calm, orderly, compassionate, and helpful to others when facing this crisis. Thus, norms
of trust and reciprocity proved to be extendable beyond traditional categories. And this
is what is needed to adapt institutional arrangements to the emergent complexity of the
world as well as to the post-demographic transition. The system of social entitlements
designed under the stable demographic perspective in the early H-phase is not fiscally
sustainable, although government debts are still held almost entirely by the Japanese
themselves. Postponing a political solution to this possible crisis of fiscal sustainability
only increases the burden on future generations. Further, as European and American
responses to the demographic transition suggest, the prospect of aging of the
population may not be an inevitable burden to society. Immigration, senior and gender
development, reversal of fertility decline,51 evolution toward a care economy, inflow as
well as outflow of foreign direct investment (FDI), and so on can not only mitigate the
problem, but may make the coming mature society livelier and richer in diversity,
although moderation in per capita income growth may be inevitable. However, in order
to make these options viable, various interest groups differentiated by the broad
categories of gender, generation, ethnicity, nationality, and such must be
accommodated and reconciled in political process. This requires a fundamental
transformation of the political institutions shaped in the heyday of the H-phase and
firmly embedded in the categorical norms of vested interests. In my view, the fact that
Japan has not yet found a practical solution to this is a fundamental reason for the
societal stagnation of the last two decades.
Perhaps I have tried to cover too many topics in my talk, for which I apologize.
However, I have just wanted to convey several simple points. On one hand, there is
commonality of development processes across economies, as the insight of the unified
approach to development predicts, so that we can mutually better understand issues
involved in development of each other. On the other hand, there are also differences in
the onset, duration, and institutional forms of developmental phases. In order to
understand this, a comparative analysis of the co-evolution of economy, demography,
and institutions is essential. I have tried to illustrate this point using the China-Japan
comparison as referential points. There can be one more point on the top of these
insights, although I do not have enough time and readiness to elaborate on this today.
That is, precisely because of differences and varieties in developmental processes, there
can be potential complementarities among developmental strategies of economies,
51

Myrskyl, Kohler and& Billari (2009) show that, using new cross-sectional and longitudinal analyses of
the total fertility rate and the human development index (HDI) of the United Nations Development
Program, a fundamental change in the demographic transition might have been occurring. Although
development continues to promote fertility decline at low and medium HDI levels, further development
can reverse the declining trend in fertility. They showed that among highly developed countries, fertility
decline may have been reversed, but that the only exceptions to this are Japan, Korea, and Canada. It is
noteworthy, however, that the decline in the total fertility rate has been recently slightly reversed in
Japan. It rose to 1.39 in 2010 after it hit 1.26 in 2005.

24

which are not possible in a homogenous world. Thus, gains from trade may not be
limited to the domain of commodity exchanges but also can be derived in the domains
of mutual flow of human beings, organizations, information, and ideas. Scholarly
exchanges among us economists also constitute a part of this process, in which the
value of a congress like this one lies. I am very much looking forward to learning from
diverse views and approaches during this week.
Thank you very much for listening.

25

References

Acemoglu, D., S. Johnson & J. A. Robinson (2001). The colonial origins of comparative
development: an empirical investigation, American Economic Review 91: 1369-1401.

Ahlering, B., &S. Deakin (2007). Labor regulation, corporate governance, and legal origin: a case
of institutional complementarities, Law and Society Review 41: 865-908.

Aoki, M. (1986). Horizontal vs. vertical information structure of the firm, American Economic
Review 76: 971-983.

Aoki, M. (2001). Toward a Comparative Institutional Analysis, MIT Press.

Aoki, M. (2010). Corporations in Evolving Diversity: Cognition, Governance and Institutions,
Oxford University Press.

Aoki, M, (2011). Institutions as cognitive media between strategic interactions and individual
beliefs, Journal of Economic Behavior & Organization 79: 20-34.

Baldwin , C. Y., & K. Clark (2000). Design Rules: The Power of Modularity, Vol. 1, MIT Press.

Becker, G. S., K. M. Murphy & R. Tamura (1990). Human capital, fertility, and economic
growth, Journal of Political Economy 98: S12-S37.

Bloom, D., & J. G. Williamson (1998). Demographic transitions and economic miracles in
emerging Asia, World Bank Economic Review 12: 419-55.

Cai, F., & M. Wang (2006). Challenge facing Chinas economic growth in its aging but not
affluent era, China and World Economy 14: 20-31.

Cai, F., & M. Wang (2008). A counterfactual analysis on unlimited surplus labor in rural china,
China and World Economy 16: 51-65.

Cao, J., M. S. Ho, D. W. Jorgenson, R. Ren, L. Sun & X. Yue (2009). Industrial and aggregate
measures of productivity growth in China, 1982 2000, Review of Income and Wealth 55,
Special Issue 1: 485-513

Chang, J. K., (1969). Industrial Development in Pre-Communist China: A Quantitative Analysis.
Chicago: Aldine.

Clark, G. (2007). A Farewell to Alms: A Brief Economic History of the World, Princeton University
Press.

Duara, P. (1988). Culture, Power, and the Modernizing State: Rural North China, 19001940,
Stanford University Press.

26


Eastman, L.E. (1988). Family, Fields and Ancestors: Constancy and Change in Chinas Social and
Economic History, 1550-1949, Oxford University Press.

Eggleston, K. N., & V. Fuchs (2011). The implications of the new demographic transition,
preliminary working paper, Stanford University.

Fehr, E., & C. F. Camerer (2007). Social neuroeconomics: The neural circuitry of social
preferences, Trends in Cognitive Sciences 11: 419-27.

Fei, H. (1939/1946). Peasant Life in China, New York: Oxford University Press

Feuerwerker, A. (1958) Chinas Early Industrialization: Sheng Hsuan-huai (1844-1916) and
Mandarin Enterprise, Harvard University Press.

Feuerwerker, A. (1980, 1983/1995). The Chinese Economy, 18701949, University of Michigan
Press (originally published in Volume 11 and 12 of the Cambridge History of China, Cambridge
University Press).

Galor, O. (2011). Unified Growth Theory, Princeton University Press.

Galor, O., & D. N. Weil (1996). The ender gap, fertility, and growth, American Economic Review
86: 374-87.

Galor, O., & D. N. Weil (2000). Population, technology, and growth: from Malthusian stagnation
to the demographic transition and beyond, American Economic Review 90:806-28.

Gold, T., D. Guthrie & D. Wank, eds (2002). Social Connections in China: Institutions, Culture,
and the Changing Nature of Guanxi, Cambridge University Press.

Goodman, L. G., N. Keyfitz, & T. Pullman (1974). Family transformation and the frequency of
various kinship relationships, Theoretical Population Biology 5: 1-27.

Greif, A. (2006). Institutions and the Path to the Modern Economy: Lessons from Medieval Trade,
Cambridge University Press.

Guinnane, T. W. (2011). The historical fertility transition: a guide for economists, Journal of
Economic Literature, forthcoming.

Hall, R., & C. Jones (1999). Why do some countries produce so much more output per worker
than others? Quarterly Journal of Economics 114: 83-116.

Hansen, G. D., & E. Prescott (2002). Malthus to Solow, American Economic Review 92: 1205-
17.

Hayami, Y., & Y. Otsuka (1993). The Economics of Contract Choice: An Agrarian Perspective,
Clarendon Press.

27

Hayashi, F., & E. Prescott (2008). The Depressing effect of agricultural institutions on the
prewar Japanese economy, Journal of Political Economy 116: 573-632.

Herrmann-Pillath, C. (2010). Social capital, Chinese style: individualism, relational collectivism
and the cultural embeddedness of the institutions-performance link, China Economic Journal 2:
325-50.

Hsiao, K. (1960). Rural China: Imperial Control in the Nineteenth Century. University of
Washington Press.

Huang, P.C.C. (1985). The Peasant Economy and Social Change in North China, Stanford
University Press.

Ishii, S. (1966).(Historical Study of the Japans State
Institutions: Power and Landownership), University of Tokyo Press

Izuma, K., D.N. Saito, & N. Sadato (2008). Processing of social and monetary rewards in the
human striatum, Neuron 58: 284-94.

Jones, C. (2001). Was an industrial revolution inevitable? Economic growth over the very long
run, Advances in Macroeconomics 1: 1-42

Jorgenson, D. (1961) The development of a dual economy, Economic Journal 71: 309-34.

Jorgenson, D. (1967). Surplus agricultural labor and the development of a dual economy,
Oxford Economic Papers, 19: pp288-312.

Kremer, M. (1993). Population growth and technological change: one million B.C. to 1990,
Quarterly Journal of Economics 108: 681-716

Kuznets, S. (1957). Quantitative aspects of the economic growth of nations II: industrial
distribution of national product and labor force, Economic Development and Cultural Change,
Supplement to 5: 3-110.

Kwon, TH. (1977), Demography of Korea: Population Change and its Components in Korea: 1250-
1966, Seoul National University Press.

Landes, D. S. (2006). Why Europe and the West? Why not China? Journal of Economic
Perspectives 20(2): 3-22.

Lee, R.D. (1988). Induced population growth and induced technological progress: their
interaction in the accelerating stage, Mathematical Population Studies 1: 265-88.

Lewis, W. A., (1954). Economic development with unlimited supplies of labour, Manchester
School 22: 139-91.

Li, H. & L. Zhou (2005). Political turnover and economic performance: the incentive role of
personnel control in China, Journal of Public Economics 89: 1743-1762.

28


Lin, J. Y. (1994). (Institutions and Technology in Chinas
Agricultural Development), Shanghai Peoples Publisher.

Lucas, R. E. (1988). On the mechanics of economic development, Journal of Monetary
Economics 22: 3-42.

Lucas, R.E. (2002). Lectures on Economic Growth, Harvard University Press.

Maddison, A. (2001, 2003). The World Economy. Volume 1: A Millennial Perspective and Volume
2: Historical Statistics, OECD Publishing.

Maddison, A. (2007). Contours of the World Economy, I-2030 AD: Essays in Macro-Economic
History, Oxford University Press.

Montague, P. R., & G. S. Berns (2002). Neural economics and the biological substrates of
evaluation, Neuron, 36: 265-84.

Muramatsu, Y. (1949/1975). (The Societal System of the
Chinese Economy), Toyo Keizai Publishers.

Muramatsu, Y. (1966). A documentary study of Chinese landlordism in the late Ching and the
early Republican Kiangnan. Bulletin of the School of Oriental and African Studies, 29: pp. 566-
99.

Muramatsu, Y. (1970). (The Land Bursary in
Modern Jiangnan: A Study of Chinese Landlordism), University of Tokyo Press.

Myrskyl, M., H. Kohler, & F. Billari (20099). Advances in development reverse fertility
declines, Nature 460/6: 741-3.

Ngai, L. R. (2004). Barriers and the transition to modern growth, Journal of Monetary
Economics 51: 1353-83.

North, D., J. J. Wallis, & B. R. Weingast (2009). Violence and Social Orders: A Conceptual
Framework for Interpreting Recorded Human History, Cambridge University Press.

Ohkawa, K., & H. Rosovsky (1973). Japanese Economic Growth: Trend Acceleration in the
Twentieth Century, Stanford University Press .

Oshima, H. (1987). Economic Growth in Monsoon Asia: A Comparative Survey, University of
Tokyo Press.

Palais, J. B. (1996). Confucian Statecraft and Korean Institutions: Yu Hyngwn and the Late
Chnson Dynasty, University of Washington Press.

Perkins, D. (1969). Agricultural Development in China: 1368-1968, Edinburgh at the University
Press.

29


Perkins, D. (1975). Growth and changing structure of Chinas twentieth-century economy, in D.
Perkins, ed., Chinas Modern Economy in Historical Perspective, Stanford University Press: 115-
161.

Pomerantz, K. (1993). The Making of a Hinterland: State, Society, and Economy in Inland North
China, 1853-1937, University of California Press.

Pomerantz, K. (2000). Great Divergence: China, Europe and the Making of the Modern World
Economy, Princeton University Press.

Qian, Y., & B. Weingast (1995). Federalism Chinese style: the political basis for economic
success in China, World Politics 48: 50-81.

Rawski, T. G. (1989). Economic Growth in Prewar China, University of California Press.

Romer, P. (19909). Endogenous technological change, Journal of Political Economy 76: S71-
S102.

Ruskola, T. (2000). Conceptualizing corporations and kinship: company law and development
theory in a Chinese perspective, Stanford Law Review 52:1598-1729.

Steinfeld, E. S. (2000). Forging Reform in China: The Fate of State-Owned Industry, Cambridge
University Press.

Tanimoto, M. (1989)
(Indigenous Economic Development and Textile Industry in Japan: Market Formation and
Family Economy), University of Nagoya Press.

Tao, R., & C. Shi (2010).-
(The contradictory theory of the Lewisian turning point and the coordinated reform of
registration-land-public finance institutions), CIDEG Policy Research Working Paper Series:
2010A0101, Tsinghua University.

Teranishi, J. (1982). (Economic Development and Finance in
Japan), University of Tokyo Press.

Tuljapurkar, S. (2010). How demography shapes individual, social, and economic transitions in
Asia, in K. Eggleston & S. Tuljapurkar, Eds. Aging Asia: The Economic and Social Implications of
Rapid Demographic Changes in China, Japan and South Korea, Brookings Institutions, 35-42. .

Watanabe, H. & F. Gomi, eds. (2002). (History of Land
Ownership), Yamakawa Publishers.

Wu, J. (2004/2005). , Far East Publishers. English translation:
Understanding and Interpreting Chinese Economic Reform, Mason Ohio: Thomson/South
Western.

30


Young, A. (2003). Gold into base metals: productivity growth in the Peoples Republic of China
during the reform period, Journal of Political Economy 111: 1220-61.

Young, L. (1999). Japan's Total Empire: Manchuria and the Culture of Wartime Imperialism,
University of California Press.

Zelin, M. (1986). The rights of tenants in Mid-Qing Sichuan: a study of land-related lawsuits in
the Baxian archives, Journal of Asian Studies: 45:499-526.

Zelin, M. (2009). The firm in early modern China, Journal of Economic Behavior and
Organization 71: 623-637.

Zelin, M., J.K. Ocko & R. Gardella, eds. (2004). Contract and Property in Early Modern China,
Stanford University Press.

Zheng, Z. (2001). Family Lineage Organization and Social Change in Ming and Qing Fujian,
University of Hawaii Press.

31