山东大学
硕士学位论文
从马尔萨斯到索罗——长期增长模型的比较
姓名：王重阳
申请学位级别：硕士
专业：政治经济学
指导教师：黄少安
20040426
原创性声明
本人郑重声明：所呈交的学位论文，是本人在导师的指导下，独
立进行研究所取得的成果。除文中已经注明引用的内容外，本论文不
包含任何其他个人或集体己经发表或撰写过的科研成果。对本文的研
究做出重要贡献的个人和集体，均已在文中以明确方式标明。本人完
全意识到本声明的法律责任由本人承担。
论文作者签名：叁垒盟
关于学位论文使用授权的声明
本人完全了解山东大学有关保留、使用学位论文的规定，同意学
校保留或向国家有关部门或机构送交论文的复印件和电子版，允许论
文被查阅和借阅；本人授权山东大学可以将本学位论文的全部或部分
内容编入有关数据库进行检索，可以采用影印、缩印或其他复制手段
保存论文和汇编本学位论文。
(保密论文在解密后应遵守此规定)
论文作者签名：至熏盟导师签名：
山东太学硕士学位论立
中文摘要
人类在地球上已经存在了犬约数万年，如果把经济增长定义为人均收入的持
续增加，用普通老百姓的人均收入作为度量经济增长的指标，则经济增长仅仅是
最近两百年左右的事情。我们可咀把这叫做现代经济增长。在此两百年之前的漫
长岁月里人类中的普通老百姓的收入，或者说生活水平基本上没有太多的变化。
我们把普通人人均收入停滞的这种状态叫做马尔萨斯陷阱，或者马尔萨斯状态，
这时历史上人类经济生活的特征。比如康超1(1986)就认为中国晚清末年的人
均收入要略低于秦始皇时普通老百姓的收入。
古典经济学家和现代经济学家都对于他们时代的经济特征用经济理论给与
了经济解释。马尔萨斯陷阱是以马尔萨斯、李嘉图为代表的古典经济学家的国民
财富增长理论的核心。在他们那里，经济增长意味着国民财富的增加．而人均收
入水平大体上是不变的，故经济增长体现为人口的增加。中国又是一个例证，数
千年灿烂文明，发达的科技水平，造就了世界上第一人口大国。而以索洛为代表
的新古典经济学家给出了现代经济增长理论，在那里普通大众的人均收入的持续
增加是最基本的事实。两个时代的经济理论都符合各自要解释的时代。那么能不
能有一个都解释历史和现代的统一的经济理论呢?能否将理论和历史统一是摆
在现代经济学家面前的任务。
在卢卡斯提出这个问题之后，许多现代经济学家郝力图用现代分析方法和理
论框架给出一个统一的模型，使之既包含马尔萨斯陷阱又能实现现代经济增长。
本文即是过王这=耋堡监地丝总结和些较。依据经济增长的动力因素是外生给定
的、至墨直搓型奎多确定的，可以将这类理论划分为外毒垡篮毽型和内生擅篓攥
型。内生增长模型中的动力可以来源于规模报酬递增的人力资本、或由人创造的
知识等。在总结和比较这些模型之后，本文得出了姐至的结论：每一种理论都只
是复建盟经遗盐盒婆塑曲二：￡；g蠛，没=鲤二尘堕型墨星鱼二切、完全!E塑的；
只有将这些理论综台起来才能呈现出一幅完全姐人类经济生意的图画；每个理论
都有它自身的假设前提，一个理论能正确的指导实践在于满足它的假设前提。
这些理论对于今日中国经济实现长期持续增长是很有启发意义的。我们可以
思考中国经济如何长期持续增长。本文对此提出的政策建议是要重视农业，提高
农业的生产率。具体地说就是降低化肥、拖拉机等农业中间品的价格，适当提高
农产品的价格。本文蚴方法是总鎏!堪较，对于每一种理论按：提超盟皿题、作者∞回
答、建堪方法和评价箜婴酆坌垄堡坌析、比较·本文不包含任何一个数学公式，
尽管所有的理论原来都是用数学模型表达的。我认为这样可以更加容易的看到一
个理论的深处，而不为表面现象所迷惑。
关键字：经济增长马尔萨斯陷阱现代经济增长
音谭，详细可见Galor＆Well(2㈣．
山东大学硕士学位论文
ABSTRACT
Human has dwelled on 1his planet for tens of thousands of years．If economic
growth is defined as sustained growth of income per capita，and the average income
of ordinary people is takep as the measure of economic growth，we will see that
economic growth is only a phenomenon of recent 200 years．Thjs is called modem
growth．Before this there had been almost no improvement in living standards．We
call this state as Malthusian trap．or Malthusian state．This was the characteristic of
human economic history．
Both classical economists and modem economists give an account of their
contemporary features．Malthusian trap is at the heart of classical economists like
Malthus and Ricardo．In their theory economic growth means the accumulation of
wealth of a nation，+and economic is embodied by the growing population．While
modem economists also have their own theories，in which the sustain improvement of
income per capita is one of the fundamental facts that all for an explanation．Both
classical and modem theories can explain their own times，and both fail to explain the
other．Can we have a unified theory that can account for the both eras and the
transition between them?
‘
After Lucas raised this problem，many economists offer their theory for it．nis
paper exposes and compares these theories．We can divide these models into
exogenous and endogenous according to whether the engine of growth is exogenous
or endogenous．11le engine of growth comes from human capital．ideas etc in
endogenous models．After the exposition and comparison，this paper arrives at the
conclusion．Every theory can only account for a portrait of the real world，and if we
want to have a complete picture of the world，we must combine them all together．
Every model has its assumptions，and if wc want to use a model to guide practice，we
should see whether the assumptions are satisfied in the specific contexts．
These theories‘are important for China today．ne policy guidance is that we
should improve the productivity of agriculture．In order to implement this we should
cut down the price of the intermediate input like fertilizer and tractors，and raise the
prices of the depreciated agricultural goods．
The methods this paper employ is to divide every theory into：the question．the
answer，the construction of the model and remarks．This paper contains rio single
equation，though all the models are expressed in mathematics．In this way we can
penetrate to the core idea behind the equations．I think and I believe in it．
Key words：economic growth，Malthusian trap，modem economic growth
2
山东大学硕士学位论文
1．Introduction
1．1 Background
I dO not see how one can look at figures like these without seeing them
as representing possibilities．Is there some action a government could
take that would lead the Indian economy to gl‘ow like Indonesia’s or
Egypt’s?If so，what exactly?If not．what is it about the nature of India
that makes it so?The COnsequences for human welfare involved in
questions like these are simply staggering：Onee one starts to think
about them，it is hard to think about anything else．『Lucas 1988，P．51
Economic growth theory is really as old as economics itself,and the famous work
of“the Wealth of A Nation”by Adam Smith is actually an issue concerning the output
of a nation．Classical economists like David Ricardo，Malthus etc，all had their theory
of production．As for modem economists who are armed with better analytical
apparatus and data，they have been more interested to give an aCcOunt of why some
countries are rich whereas others poor and tried to ofief policy guidance．Though
human beings have。dwelled on this planet for thousands of thousands of years，it is
only the recent years that man discovers the secrets of nature and utilizes these to
accumulate affluent materials that ordinary people can enjoy．Especially after World
War Two．people are amazed at the rapid growth of the economy and try to provide
satisfactory answers as many as possible．Just as the words said by hcas，the growth
theory is so fascinating and challenging that alI other issues in economics are minor
compared with it in a way．
Growth theory concentrates on the relative long term growth of economic
products，in contrast with fluctuation theory which is engaged mainly in short term up
and down of outputs．Here‘‘growth”refers to the improvement of income per capita,
or products or outputs per people，or living standards，all of which have the same
meaning in this Paper．And we are not interested in the comparison of living standards
of ancient emperor like Qin Shihuang with nowadays-rich people like B丑l Gates．
Thus when we talk about economic growth，we mean the improvement of living
standards of ordinary people，which is emphasized in Lucas 1998．And though
personal effort is the direct force that drives an economy,in modem economic growtb
theory,we generally analyze 3 basic factors of production：technology，population，
and capital including physical capital and human capital，becaUSe in macroeeonomics
we can only concentrate on the variables that can be aggregated．In some cases we
add the 4”’faCtor：natural resources。Iike Iands．Almost all economists agree that it is
these factors that work behind the growth of an eeonom孔but how they interact with
each other has been and will probably be under debates among economists．
Some caveats are in order．What is the index￡o measure economic growth?Total
GDP,or average GDP?GDP,or GNP?Thoulgh we often USe,the growth rate of total
GDP to show economic growth，the more general way is to use average，namely per
person index to measure growth．The di自fefence between GDP and GNP comes from
the economic activides between CORntries．As part of the common setting of the
3
山东大学硕士学位论文
models，which will be stated clearly in section 3，we only care for close—economy
models．Thus there exists no difference between GDP and GNP in the models exposed
in this paper．
Why to choose this topic?
AIl the theories this paper reviews a／e about economic growth in the longest
period：the economic history itself．Ever since the time history was recorded，the
number of people in the world and the products they produced grew at roughly equal，
slowly increasing rates。．Amazingly the living standards of ordinary people in the 18”
century in Europe were about the salne as those of people in China at that times or
ancient Rome or,indeed as those of people in the poorest countries in the world todav．
Or we call this long period as Malthusian state．or Malthusian trap．Then during也e
following 200 years，both production and population growth have accelerated
dramatically and production has began to grow much more rapidly than population．
For the first time in histo阻the living standards of masses of ordinary people have
begun to undertake sustained growth．It is really a wonder that a human society has
generated sustained improvement in the material aspects of the lives of all of its
members，not iust those of a ruling class，Mankind has entered all entirely new era in
his economic history．Can we give an account of this faet?What ale the
opportunities and necessities a poor country can make use of to catch up with the
rich?To answer these questions we need a theory to organize these fac纽to
explain them and to judge opportunities and necessities．Really this is one of the
most significant research chaUenges facing economists interested in growth and
development．Many economists have offered their theories that can explain this fact．
And this paper tries to expose and compare these theories．ne contribution of this
paper not only lies its exposition of the theories,but also in its enlightening
economic explanation of why China feU behind other nations and what steps we
should take to catch up with others．In brief this paper exIx)ses unified theories of
growth and transition．and states its implicatiou for China today．
t ne title“from Malthus to Solow"’deserves some explanation．Here Malthus
stands for Malthusian trap and the classical theories that tried to explain it．and Solow
stands for modem growth and the neoclassical the．odes曲at tries to explain it．珏Ⅱs
the title means both the transitiou in history and a unified theory也at tries to ac．cotmt
for this transition．
As we win see there are many different theories也at can model the transition in
different ways，and one Call’t help wondering why them are so many different models
for the sanlc phenomenon?How can they give different answers to the sanle questions?
And which one is fight,or more reasonable?Now let’s digress into thCSC questions，
because it can enhance our understanding of all the models and give better policy
guidance，which is more important，for l think the ultimate goal of a theory is to guide
practice．
2 all facts mentioned following will be present in section 2 quantitatively
4
山东大学硕士学位论文
ne purpose of science is not to veriey something，but rather to falsiry something，
especially when it comes to social science．Unlike natural ones，social phenomena are
overwhelmin酿Y complex，and there has never existed a unified theory that can
account for a11 aspects of all social events．Even the thought of finding this theory is
impossible．One theory can only explain 1imited aspects of a thing．Diversified
theories should be provided to explain the diversified world．All the theories can
explain something in limited aspects and only this．Thus how carl we compare
different theories?Purpose determines theories．The ultimate aim of giving a theory is
t0 give policy guidance．And we should compare different theories and see which one
is logically reasonable and agrees well with the observation of real world．This
criterion does not seem too surprising，but actually many'even a generation of
economists have omitted it．Modem growth theory originated in 1960s，and once
declined in 1970s．Part of the causes is that they provide all kirids of models，just from
theoretical assumption and no real grounds，and thus deserted by people in the end．
This paper tries to use this criterion and judges which theory of transition is
more suitable for China．
1．2 the Theoretical Significance in Modeling the T|ansition
None of the classical economists，like Malthus and Ricardo，forecasted the later
SUStaincd growth．In their theories the central problem is to account for the tendency
of income per capita in any society to be fixed to a roughly constant．stable leve】in
the face of improvements in technology．T110ugh they did not deny that this stable
income level might differ from one economy to another,meir theory attributed such
differences to differences in preference or CUStom。neither in technology uor in
available resources．Differences in societies’ability to produce would then induce
differences in population only．
Modem theories of eo_ⅪHomic growth，including the Solow model，have similarly
failed to deal adequately with the phenomenon of stable income and growing
population with increasing technology．These theories are built around a positive rate
technolcIgical change．either assumed or generated as an equilibrium outcome by the
assumption constant or increasing returns to the accumulation of knowledge．They are
made consistent with the sustained growth of incomes per capita by the even more
central assumption of a舀ven，constant rate of population growth．In contrast to the
classical prediction that technological change affects only changes in population，
modem growth theories generally assume that technolo舀cal change affects only
income，with no effect on population．
Both elassical and modem theories of production thus succeeded in accounting for
the central features of the contemIporary behavior of production and population．Yet
both of them failed to explain the data the other was designed to explain．Thus we
need to discover a more general theory in which the two growth thcories can be seen
as two special cases，a theory that shows the nature of the transition from the situation
of stable income that has characterized most of history to也e sustained growth that
has emerged in the Iast two centuries．And what is more important is that in the roodel
the economy should be able to transit from Malthusian state to modem growth
5
山东大学硕士学位论文
endogenously．Growth economists have ofiers many such theories and this paper tries
to exPOse some of them，which is the first reason why I think this Paper is meaningful．
Secondly these theorics can help explMn the cross-country income differences
today．Whv some countries are rich whereas others poor today?Why is there such a
big gap in living standards between developed countries and developing countries?As
recently pointed by Lucas(1998 and 2000)，the timing of a courttries transition
directly affects the country’s position in the world income distribution today．Some
countries have undertaken the accelerating modem growth trend，some are trying to
catch up with it，and some are still in the sta辨ant Malthusian state．Thus those who
have enjoyed the 200 years’modem growth are rich，those who arc catching up are
relatively rich，and those still in Malthusian trap are poor of course．
Really the incentive to write this paper derives directly from Lucas's penetrating
papen The basic blocks of this paper are influenced by Lucas’papen
This papler tries to presents all the theories that model the transition from
Malthusian trap to modem growth and the rest of the paper is organized as follows．
Before the exposition of lhe theories，section 2 presents the styled facts the
introduction mentioned above in details．In section 3山e comnlon basic assumptions
and analytical methods are explained which is embodied in the setting of all the
models presented in this paper．I divide all the models into two classes according to
general growth theory：endogenous models and exogenous models3．Section 4
summarized endogenous models and section 5 summaries exogenous models．Section
6 concludes by summing up all the models．Section 7 giyes也e policy guidance，
which I think is suitable for the current situations in China．
2．The Basic Facts’
This section provides a quantitative description of the facts mentioned above，Of
course some data arc extrapolated，but since our aim is to illustrate the general facts，
this doesn’t make too much difiefence．
2．1 Wbrld Population and Production
Figure 1 plots world population and the total world production of goods and
services，namely GDP,from the year 1000 up to now．A logarithmic scale is used in
this figure so that one can see clearly也e acceleration in both of these series．The scale
on the left is millions of PCI'sons and billions of 1985 U．S．dollars．Populatitn from
earlier years is from McEvedy and Jones(197鼬．Lucas said，earlier production
estimates are extrapolationg based on the assumption that income per capita was equal
to their 1750 levels in all—earlier years．Thus prior to 1750 the total population and
production series differ by a constant．
The estimated population for the year 1000 is 265 million．By 1960．world
3 whether a model is exogenous or endogenous depends on whether the engine of
growth in the model is exogenous given or decided in the model itself．
’this section rgfe幅to and话based oll Lueas 1998
6
些查查兰堡主兰垡堡兰——
population had grown up to 3 billion；by 1990 it was 5．2 billion．Lucas said that one
could see the acceleration of population growth since 1800 On the figure．In a world
with constant income per capita’population growth is itself a measure of production
growth．and hence according to classical theory a measure of technolc’毋cal change·
Thus whatever it Was that’was new to economic life in the 18。century,it Was not
technological change，for continued improvement In technology was needed tO
support the impressive population growth in all the centuries proceeding，as Lucas
said．
Our knowledge of production and living standards at different places and times
haS grown enormously in the past few decades．The most recent empirical
contribution is the Penn wbrld 7rable project conducted by Robert Summers and Alan
Heston．It organized data set containing population and production data on every
country in the world，from 1950 or 1960 to the present time．Data on real productiol,
are converted to common units on the“purchasing power parity’’basis that 1S
consistent with index number theory．
As a result of the Penn project，Imcas said，we now have for the first time a
reliable picture of production in the entire world，both rich and poor countries．Let 7S
review the main features of this picture，beginning with population estimates-Over
the 30 years period from 1960 throuIgh 1990，world population grew from about 3
billion tO 5．2 billiOn，or at all annual rate of 1．8 percentages．
There is a lot of poverty and starvation in this world，but actuaUy poverty is not
increasing．Over the same period during which population has grown from 3 billion to
5．2 billion．total world production grew much faster than population，from$6．7
trillion in 1960 tO$22．3 trillion in 1990．That is，world production more than tripled
over this 30一year period。growing at all annual rate of 4 percents·Production per
pemon grew at 2．2 percent per year,which is to say that the living standard of the
average world citizen nearly doubled．Thus we can safely arrive at the conclusion that
the entire human race iS getting rich，at historically unprecedented rates．
Figure 1：World Population and
Produotlon
—曲=} 。气左甥
√，。广
i
要。邑jI—{星蓑jlj量
山东大学硕士学位论文
2．2 Income Distribution
Average figures like these mask a lot of diversity．Figure 2 is one way tO use the
information in the Penn Wbrld T抽le to Sulnnlarize the distribution of the levels and
growth rates of population and per capita incomes in the post war world．The horizon
axis iS GDP in thousands of．1985 U．S．dollars on a logarithmic scale．The vertical axis
is population．And the volumes stand for population
One can see from figure 2 that the number of poor COUP,tries with mean incomes
below$1100 has declined between 1960 and 1990．On the other hand，the degree of
inequality is enormous．
Whether per capita incomes arc conve呼ng to a common(growin酌level，or
possibly diverging is impoaant．From figure 2 it is evident that this is a fairly subtle
question．and there is no definite condusion．
Figure 2：Income Distribution
2．3 GDP per capita．Fivb Regions 1750．-2伽0
For years much before 1960 there is no single source for production data for all
the world’S countries，and indeed，most of the world’S current nations did not even
exist aS independent states"prior to World W缸Two．But for a few economies⋯．rich
countries⋯．excellent national product data are available extending way back into the
nineteenth centum确e per capita rcal GDPestimates ale constmeted according to the
table 2 in the data appendix．
8
山东大学硕士学位论文
For poor societies we often have no national product data，or data of very lOW
r“ability,even for the postwar years．Lucas said his estimates reported in Table 2 for
Aflica and much of Asia for the years before 1960 are in large part extrapolations
back in time of the 1960 estimates provided by Summers and Heston．Lucas said
since in today poor agricultui'al countries the societies have been stagnant and ways of
production alSO dormant，thUS incomes levels cannot change too much from ancient
time to nowadays．As Lucas said，though we do not have sophisticated national
income and product accounts for countries Iike Korea 100 years ago，we do not need
them to arrive at fairly good estimates of living standards that prevailed back then．
"Iraditional agricultural societies are like one another,Lucas said，all over the world
and over time，and the standard of living they yield is not hard to estimate reliably．
Figure 3 illustrates the uneven C,OLlrse of the industrial revolution in different parts
of the world．Group 1 is those in which per capita incomes first exhibited sustained
growth。Group 3 consists of the rest of northwest Europe，the countries that began
sustained growth somewhat later．Group 4 is the rest of Europe，together with the
European dominated economies of Latin Africa．Group 2 is Japan，isolated only
because of its remarkable economic history,Lucas said．Group 5 contains the rest of
Asia and Africa．
Figure 3：GDP Per Capita。Rve
RegiOns
。
’B∞O
1翻mO
呈1锄
吕9000
警产
e_D00
3啪
O
l
ii
／笈／ ／
—．—，易留．
誉
1啪Populmlan
In mI咐orJI
I UK嘲，Canada．Austrlt|lL NowZDstand 384
II Japan ． 124
嘲FraNce．Gwml吖，tNNrtlterle,clu，SⅪandlnmrla 粥
Ⅳ RestofWeetlrnEmpe．LmJnAmerica．847
Eluaarn Europe。Soviet Union
V^●k佃啪P|Jail珥巾．^f一％ 筠∞
9
山东大学硕士学位论文
As shown in figure 3，per capita incomes were approximately constant，over space
and time，over the pedod 1750—1800．From figure 3 we can see that the inequality
among societies that figure 2 displays，is a recent event，emerging for the first time in
the 19“century and reaching current levels only in the 20th century．This reinforces
our former statements．
2．4 Demographic Transition
As hcas said it is hopeless to try to account for income growth since 1800 as a
purely technological event．Technological change occurred rapidly after 1800，but it
occurred at all accelerating rate of centuries prior to 1800 as well．Lucas said，what
occurred around 1800 that is new．that difierentiates the modem age from all previous
periods，is not technological change by itself but the fact that fertility increases ceased
to translate improvements in technology into population．Lucas points out that the
sustained growth is invariably associated with the reduction in fertility known as the
demographic transition。
Figure 4：Demographic Transitions
乱入：
人
股⋯飞么{＼
7∥I o_‘、、慕—。?--· _，’’一一
■--’- ：
O 2 4 6 8 10 12 14 16
Per Capita GDP．Thousands of 1985 Dollars
2·5 Summary ．
As Lucas mentioned，the most striking feature of the data reviewed in this chapter
is the fact that only two types of aggregative behavior are observed，one with per
capita income roughly constant at a level in the range of$400 to$800 1985 U．S．，and
another with per capita income growing at rates in excess一_of one percent per year．
This observation is the basis of all the models described in this pal：leL
3．Basic Setting of Modern Growth Theory
The term‘‘theory”is used in so many different ways and with different meanings
lO
5
2
S
1
5
O
5
2
1
O
O
譬薅正暑蓍2a
co—瓮In厶O儿
山东大学硕士学位论文
that we may mistake each other if we do not defino its meaning before we consider it
seriously．Briefly a theory is model whose program can bc run on a computer,This is
given in LuCaS 1988 and I think theU吐it seems too narrow a dcf'mition，but it can
really help US grasp the core of a theory acourately and efficiently,and can be tested
with data．These two points，especially the second one，make all other possible
definitions of theory little．ThUS from now on the term“model”is used with the same
meaninE as“theory”．
When we talk about modem theory'actually we mean neoclassical models．AIl the
models this PaDer exposes are neoclassical models．And they share some common
settings of in the frameworks of these models．
First，optimising tradition．
In neoclassical models，there are utilities．maximizing Consumers and
profit—maximizing profit firms．And agents namely consumers and firms are rational
expectation，which is the approximate aggregate behaviour of many agents．
Second，close economy．
In all the models exposed in this paper there iS only one economy and no
interaction between nations．In history there was little or no trade between nations．
Thus close economy assumption is appropriate．But when it comes to world today,
where countries are SO correlated tightIy by globalisation that close economy
assumption seems odd．I think new models maybe overcome this drawback．
ne last．on equilibrium pa
Much of growth theory is about the characteristics of steady states．The steady
state caIl be thought as an evolution along which OUtput and the stock of capital grow
at the same constant．ne importance of equilibrium analysis in growth theory has
both theoretical and empirical roots．Most growth models have at Ieast one stable
steady state and hence it iS a natural object of attention．What’S more important is that
the advanced industrial economies may be converging to steady state．
4．Endogenous ModeI of Growth
Since our attention iS on production and population and we wailt to find a unified
theory of production and population，the fertility choice should be endogenized in
models．So the First model We：think of iS the model in which most factors are
endogenous．Thus I expose endogenous models first．Generally speaking endogenous
models are those in which the engine if growth in determined in the model．which iS
usually human capital or non—rival knowledge，or just assumed relationship between
producing factors．’
4．1 Human capital based model
4．I．I the Question
Though they were not armed with modem analytical methods and macro data
processing methods，classical economists like Adam Smith and David Ricarrie had
actively debated over a vast amount of historical information about the levels and
rates of growth of living standards．No one mentioned sustained growth among
山东大学硕士学位论文
classical economists．Smith and Ricardo took it for granted that the wealth of a nation
meant their incomc levels and not their growth rates，and it was the stability of these
levcls over time that should be cxplained．Thas when Malthus ofiered a fertility
model that could ex．plain the observed stability in living standards．other economists
of his time accepted it without hesitation．
And as for modem economists，the sustained growth is a thing that goes without
saying．Solow proposed a theory of sustained growth，which is designed to fit the data
of the 200 years history of U．S．Thus classical economists gave the theory consistent
with their times and modem economists have their own models agrees with modem
world．
Can wc have a unified model that can account for both history and today?Can
there be a transition from Malthusian trap to modem growth in the model?
"I tlese are the central question that Lucas 1998 addresses．And it is also the central
question that all other theories in this Paper try to answer．
4．1．2 Lucas’the Answer to the Question
The central idea in Lucas’papcr is that the SuCCeSSful transformation firom an
economy of traditional agriculture to a modem，growing economy depends crucially
on an increase in the rate of accumulation of human capital．Societies and their
citizens must be open to the“new possibilities that development creates．
Lucas views the sustained growth associated with demographic change······the
reduction in fertility．He considers these two phenomena as the two sides of a sarne
thing．As Lucas meritioned above technological progress itself calmot account for the
transition around 1800．And as already demonstrated in neoclassical model that
physical capitals cannot sustain income growth due to decreasing return to scales．
Thus human capital is the ri曲t one to explain the sustained growth．Human capital
can be the eugine of growth、which is already shown in Lucas 1988．And he assumed
that a household faces the problem of quantity—quality choice．This means that if a
house wants to have more children then every child will get less education and thus
accumulated less human capital．And if a household likes to have fewer children．
every child can get better education and have more human capital．Thus when the rate
of accumulation of human capital is low,a household will adopt few children．and if
the rate of aCCumulation of human capital increases。a houseJ]Iold tends to raise fewer
children and provides them with better education and thus more human capital．Thus
iIl Malthusian state the rate of aecumulation of human capital is low and households
dedde to raise children with 1ittle human capital．At this time if technology pro琢esses
and the rate of human capital accumulation is still low,a household will adopt more
children but gives every chfldren low human capital．Thus we have a world where the
living standards tare roughly constant and technology advances bring about more
population．This is fight the Malthusian state．If the rate of accumulation of human
capital increases to certain extent and a household decides to have less children and
provides them with more human capital，the fertility rate drops，thus we can have less
fertility rate．This models the transition from Malthusian state．And the left task to
model sustained growth is easv．The key point above is the change of rate of
accumulation of human capital．
12
山东大学硕士学位论文
4．1．3 the Construction of the Model
Lugas gives mainly two models．The first one is a modem，mathematically explicit，
version of Ricardo’s theory of production and distribution for all economy with fixed
1and and labor as factors of production．In another modeI human capital is introduced
and modem sustained growth is generated，where Lucas shows that exogenous
technolo西cal change cannot be made consistent with the demographic transition．
4．1．4 Remarks
Lucas is the first to stimulate the research interest in the transition from
Malthusian state to sustained growth．He associates this transition with demographic
change．Influence by others，he 2iyes a roodel in which the main problem a household
faces is a quality．quantity choice．Thus the model has bom Malthusian state and
sustained growth，according to different rate of accumulation of capital．But why the
rate of accumulation of human capital is difierent?Lucas defines it!This is the big
drawback of the model．If there were rio increase in rate of accumulation in human
capital，there would not be the transition of course，according to Lucas’paper．
This model is the benclunark of all the following models．This roodel gives the
basic facts and the ri曲t question，and also leaves an important open question：how to
model the transition?砧l the following models try to solve this question．
4．2 Idea Based Model
4．2，1 the Question
For thousands of years，the average standard of living seems to have risen very
little．despite increases in the level of technology and large increases in the level of
the population．Then，after thousands of years of little change，the level of per capita
consumption increased dramatically in less than two centuries．
Can we have a growth model that is able to match several key facts of this
economic history?
Jones writes a paper titled“Was an Industrial Revolution Inevitable?
⋯⋯Economic growth over the very long run”and tries to address this problem．
4．2．2 the Answer to the Question
Jones sets uD a quantita(Ne model and there arc tWO factors that are central to
understanding this history．The first is a viauous circle：more people produce more
ideas．and more ideas make additional population growth possible in turn．The second
is all improvement in institutions that promote innovation，such as property ri#ats，and
the simulated economy indicates that the most important single factor in the transition
to modem growth is the increase in the fraction of output paid to compensate
inventors for the fruits of their labor．
What is more important．the quantitative analysis in the model suggests that
changes in institutions to support innovation have been extremely important，Not only
the rise and decline of institutions such as property ri曲ts could be responsible for the
rise and decline of great civilizations in the past，but also the establishment of
innovation—promoting institutions in the 20th century appears to have played a critical
role in generating the observed Industrial Revolution．
In brief Jones model says that more people more knowledge and more knowledge
山东大学硕士学位论文
more growth，hence more people．And institution plays all important role in the
transition in the simulated economy．
4．2．3 the Construction of the Model
Jones 1999 sets up all idea-based theory of growth in which people are a key input
into the production of new ideas with a model of endo·genous fertility and mortality in
order to analyze those remarkable facts．In his model fertility is chosen in a utility
maximizing framework．ne model is an Over Lapping Generation(OLG)model．
ne economy consists of Nt identical individuals，where t=0，1，2，⋯indexes
time．Each individual obtains utility from consumption and from tlle number of
children produced by the individual in each period．Individuals are each endowed
with one unit of labor per period，which they can use to obtain consumption or to
produce children．
The production function is assumed to exhibit constant returns to scale to the rival
inputs of labor and land，and therefore increasing returns to labor，land，and
knowledge taken together．．as in Romer(1990)，this assumption reflects a key
property possessed by knowledge．Knowledge is nonrivalrous and Call therefore be
used at any scale of production without having to be reinvented．
The intemal dynamics provided by the model are able to produce thousands of
years of virtually no sustained growth in standards of living despite increases in both
technology and population，followed by the emergence of rapid growth．More
generally the model matches the broad time series behavior of both population and
per capita consumption．To match the population data exactly，however，the model
introduces two shocks．
The first shock is all improvement in property fights．The fraction of output that is
allocated to compensate inventive effort is an exogenous variable in this model．In an
economy with a well—functioning system of property rights，inventors are allowed to
earn the retums from their discoveries，either throu庐some direct mechanism such as
a prize or throvIgh all alternative mechanism such as the monopoly rents that acelne to
the owner of a patent．Intellectual property rights are obviously important in such a
system，but so are more general kinds of property rights：a patent is valuable only
insofar as the owner is allowed to recoup one—time invention costs through subsequent
sales of some product．
ne second shock introduced in the model is a temporary decline in the standard
Solow(1956)measure of total factor productivity，as might OCCUr during times of war
or famine．
What should be emphasized here is that the quantitative analysis of the model
assigns a major role to changes in property rights in explaining growth over the very
long run．Otherwise there will be no transition in the model．
4．2．4 Remarks
There is not too much novelty in the way that Jones models the linkage between
populations and discover of new ideas．Kremer 1993 and Lee 1998 have a similar
model．And Jones is not the first one to emphasizes the iraportance of institution such
as property rights：North and Thomas(1973)，Rosenberg and Birdzell(1986)，Jones
14
山东大学硕士学位论文
(1988)，and Mokyr(1990)suggest that over the broad course of history，changes in
such property rights have been a fundamental determinant of economic growth．The
contribution of this paper lies in its emphasis on quantitative theory，i．e．in providing a
complete quantitative analysis of the growth model．In a quantitative method this
paper reconfirms the importance of institutions like property right．
The most obvious counter example is old China，who had a large population and
defeated by Europeans．More people and more population is only part of the story，
and many other policies and institutions matter,I think．
4．3 the ModeI of Population．Technology．and Growth
4．3．1 the Question
Galor and Weil divide history into three re西mes according to the historical
evolution of population，tecInlology，and standards of living．It develops a unifled
model that involves the transition between three distinct regimes that have
characterized the process of economic development：the fwst is a Malthusian regime，
where technological progress is slow and population growth prevents any sustained
rise in income per capita，the second is a Post—Malthusian re西me，where teehnological
progress rises and population growth absorbs only part of output growth，and the third
is the modern growth with sustained growth and almost stagnation of population
growth．Galor and WeiI give historical evidence suggests that the key event that
separates the Malthusian and Post．Malthusian Regimes is the acceleration in the pace
of technological progress，whereas the event that separates the Post．Malthusian and
Modern Growth eras is the demographic transition that foUowed the industrial
revolution． As Galor and Weil point out the efflergence from the Malthusian trap
and the onset of the demographic transition raise intriguing questions．
How was the link between income per capita and population growth．which had
for so long been a constant of human existence，so dramatically severed?How does
one account for the sudden spurt in growth rates?Is there a unified framework of
analysis that can account for this intricate evolution of population，technology，and
growth throughout human history?These are questions their paper addresses．
4．3．2 the Answer to the Question
In their model，parents switch out of quantity and into quality，but do so not in
response to the level of income but rather in response to technological progress．As
the paper shows the“disequilibria’’brought about by technological change raises the
rate of return to human capital，and thus jnduccs the substitution of quality for
quantity．At the heart of their model is a novel explanation for the reduction in fertility
that has allowed income per capita to rise so far above subsistence．And the choice of
parents regarding the education level of their children affects the speed of
technological progress．ne model also links the size of the population to the rate of
technological progress and to the take-off from the Malthusian Regime．The fmal
piece of the modelis the most Classical．The economy is characterized by the
existenceofafixedfactorofproduction，land，and a subsistencelevelofconsumption
below，which individuals cannot survive．And shocks to the ratio of land to labor will
山东大学硕士学位论文
induce temporary changes in the real wage and fertility，which will in turn drive
income per capita back to its stationary equilibrium level．Because technological
progress is slow，the return to human capital is low，and parents have little incentive
to substitute child quality for quantity．The Malthusian pseudo steady state vanishes in
the long mn because of the impact of population size oil the rate of technological
progress．At a sufficiently high level of population，the rate of population-induced
technological progress is high enough that parents find it optimal to provide their
children with some human capital．At this point，a virtuous circle develops：higher
human capital raises technological progress，which in turn raises the value of human
capital．Increased technological progress initially has two effects on population
growth．On the one hand，improved technology eases households’budget constraints，
allowing them to spend more resources on raising children。On the other hand，it
induces a reaUocation of these increased resources toward child quality．In the
Post—Malthusian Regime，the former effect dominates，and so population growth rises．
Eventually，however,more rapid technological progress resulting from the increase in
the level of human capital triggers a demographic transition：wages and the return to
child quality continue to rise，the shift away from child quantity becomes more
significant，and population growth declines．In the Modem Growth Regime，
technology and output per capita increase rapidly，whereas population growth is
moderatc．
4．33 the Construction of the Model
From the model we c驵see that differences between countries in the
determination of population growth or in the process of technological change would
be reflected in their ability to escape the Malthusian trap／md in the speed of their
takeoff．And the differences are generally attributed to institutions and cultural factors
Similarly,differen∞s in policies，such as the public provision of education，would
change the dynamics of the model．
One interesting possibifity that the model suggests is that the inflow of grain and
other commodities as well鸱the outnow of migrants during the nineteenth century
may have played a crucial role in Europe’s development，aS Galor and Weil
emphasizes．By easing the land constraint at a crucial point--when income per capita
had begun to rise rapidly，but before the demographic transition had gotten under
wa)“·the“ghost acres”of the New、Ⅳo订d provided a window of time，which allowed
Europe to pull decisively away from the Malthusian equilibrium(Kenneth Pomeranz，
1999)． 。
4．3．4 Remarks
This modelis relative complex compared with others．Its main drawbacks are that
there is no simulation in the model．What’s more．endogenous models are intrinsically
lack of convincing force．Many relationships between variables are just assumed．
Many meanings behind a varia务le are not clear．髓Us we call concert to exogenous
models．
5．Exogenous Model of Growth
16
山东大学硕士学位论文
Exogenous models are those in which the engine of growth is exogenous．Often
population growth is assumed to be related with consumption or output per person．
Technology progress is exogenous，and it is both the key exogenous variable and
engine of growth·
5．1 the Model of Exogenous Growth ofTo组l Factor Productivity crFP)
5．1．1 the Question
Can we have a unified growth theory that accounts for the roughly constant living
standards displayed by world economies prior to 1800 as well as the growing living
standards exhibited by modem industrial economies?Was the industrial revolution
inevitable．which is the transition from an era when per capita incomes are stagnant to
one with sustained growth?
5．1-2 the Answer to the Question
In their model the transition is inevitable given positive rates of total factor
productivity growth．They use a standard growth model with one good and two
available technolo—gies．The first，denoted as the Malthus technology,requires land，
labor'and reproducible capital as inputs．The second，denoted as the Solow
technology,does not require land．It is showed that in the early stages of development，
only the Malthus technology is used and，due to population growth，living standards
are stagnant despite technologl‘cal progress．Eventually,technolo舀cal progress causes
the Solow technology to become profitable and both technologies are employed．At
thi§point．1iving standards improve since population growth has less influence on per
capita income growth．In the limit，the economy behaves like a standard Solow
growth model．
In particular,their theory predicts that land’s share in production should faU
endogenously over time，as observed historically，and that there will be an escape
from Malthusian stagnadon and a transition to modem growth in the sense of Solow．
What’s more this transition from constant的growing living standards is inevitable
given exogenous positive rates of total factor productivity growth and involves no
change in the structure of the eeonom)／，namely the parameters describing preferences
technology，and policy etc．
5．1．3 the Construction of the Model
Thjs is a one-good．two·sector overlapping-generations(OLO)model．In the first
production sector,which we call the Malthus sector，capital，labor，and 1and arc
combined to produce output．In the second seetor．which we call the Solow sector,
only capital and labor are used to produce the salne good．TOtal factor productivity is
assumed to be eXOgeUOUS to these individual profit centers．This is one of the key
assumptions of the血odel．Output from either sector can be used for consumption or
investment in capital．Capital is assumed to depreciate fully at the end of each period．
Households live for two periods and have preferenCcs that depend on consumption
in each period of life．To follow n10mas Malthus(1798)．and也e more recent work of
Kremer(1993)，it is assumed that the population growth rate depends on the living
standard，which measured by．using consumption of a young household．In addition，it
assumed that this funcdon continuous，differentiable，and single-peaked．
17
山东大学硕士学位论文
Givcn the initial values for population，capital stock，and land，a competitive
equilibrium in this economy consists of sequences of prices，firm allocations and
household allocations．And the equilibrium is characterized by the following results：
For any wage rate and capital rental rate．it is profitable to operate the Malthus sector．
And Given a wage rate and capital rental rate,maximized profit per unit of output in
the Solow sector is positive if and only if the T0tal Factor Productivity in the Solow
sector is above certain level．Thus Both the Malthus and Solow sectors will be
operated if and 0111Y if the TFP in Solow seetor is above certain level．
They choose initial conditions and the俯sequcnces in beth sectors SO that the
economy is initially using only the Malthus technology and then study how the
economy develops over timc．A sufficient conditions guaranteeing that the Solow
technology will eventually be adopted is stated．and in order to assure that there is a
transition to the Solow economy with land being of minor importance in production，
the paramelers of the model is restricted．
Thcn a quantitative exercise is carried out．The economy is inifially in a
Malthusian steady state．and then they simulate the equilibrium path until essentially
all the available capital and labor are employed in the Selow sector．In the simulation
the growth rate of TFP is assumed to be fixed as a constant which is bigger than 1．
And the model is so caUbrated that(1)the initial Malthusian era is consistent埘tll the
growth facts describing the En醇ish economy prior to 1800，(2)the Solow-only
economy matches也e growth facts describing post．World WarⅡindustrialized
economies，(3)the population growth rate reacts t0 changing living standards as
reported in Kremer(1993)and Lucas(1998)，and(4)the implied annual rate ofreturn
on capital is reasonable giyen available data．In addition，they use data in Lucas(1998)
on population growth rates and per capita GNP for various regions of the world from
1750 to the present to calibrate the population growth function．which is the function
of consumption．Population growth rates appear t0 increase linearly in living
standards from the Malthusian level to the level where population is doubling each
period(every 35 years)．Over this range，living standards double from the Malthusian
level，After this．the population growth rate decreases linearly until living standards
are approximately 18 times·what they were in the Malthus steady state．It is assumed
that population is constant as consumption grows beyond this point．Finally the
simulated growth path has several featurcs in common with the historical data of
England．During the periods when only the Malthus technology is being used，
population grows at the same rate as output,and the wage stays constant．After period
0，population growth increases，and the real wage increases as well，where the wage
has been normalized to equal one in the Malthus steady state．
5．1．4 Remarks
11ljs paper medcls both the Malthusian trap and modem growth，and the transition
between them，using exogenous growth technology．In particular,this model predicts
that“land’s share in producdon should fall endogenously over time．as observed
historically，and that there will be an escape from Malthusian stagnation and a
transition to modem growth in the sense of Solow．”The emphasis of也e role of land
in production and in transition is sereinal in this Paper。which is one of the reasons
18
山东大学硕士学位论文
that make this paper great．
Compared with endogenous growth model，this model may appear too simple．But
iust as Hansen and Prescott points out that‘'we do not believe tllat there yet exists a
theory of knowledge accumulation with the same level of acceptance that is ac00rded
to the standard theory ofcapital accumulation．’’Here it is worth a digressing into the
comparison of exogenous and endogenous growth models．
Generally speaking there are two reasons that exogenous growth model is
preferred to endogenous one．The first is a theoretical one：in endogenous models the
source of growth comes from non．decreasing returns to scales，and if this is the case
why don’t we have just one firm that aggregates all the other firms?ne second is
concerning empirical work．ne famous paper is Jones．1995．He points out that
according to endogenous growth theory,permanent changes in certain policy
variables have permanent effects oo the rate of economic growth．Empirically,
however,U．S．growth rates exhibit no large persistent changes．Therefore，the
determinants of long-mn growth hi曲li曲ted by a specific growth model must
similarly exhibit no large persistent changes or the persistent movement in these
variables must be offsetting．Otherwise，the growth model is inconsistent with time
series evidence．So Jones 1995 argues that many AK—style models and R＆D．based
models of endogenous growth are rejected by this criteflon，and the rejection of the
R&D—based models is particularly strong．
And there are more reasons for exogenous model．Really technology progress is
partly exogenous and partly endogenous．But most of the big innovations are created
by genius，and thus we Call consider the progress of technology is exogenous．And as
for this paper,they are not addressing the problem why technology progresses，but
rather that why people don’t use advanced technology,which is actually what
happened in history．
Thus．in a11．exogenous model is preferred to endogenous one．And as Hansen and
Prescott emphasizes that endogenous growth features call be easily incorporated into
their theory in a way that does not alter their main findings．
Actually Hansen and Prescott’s theory is silent about why population growth rates
are increasing in living standards in the early stages of development and then become
decreasing in living standards at more advanced stages．Really if we speak for the
sake of simulating the model and checking whether it agrees with history．this doesn’t
matter．But as Lucas emphasizes that quality—quantity choice is an important issue in
growth and human capital is one of the engines of growth．Thus endogenizing
population growth is and should be a promising work．Thus the simple giyen
relationship between consumption and population growth is the drawbacks of this
paper,where improvement should be expected． ‘
In this model productivity growth or technological change outside agriculture is
the driving force for the transition from Malthusian trap to modem growth．One
implication of this model is that productivity increases in Malthusian sector would
actually delay rather than speed up the transition．However，as Zhu Xiaodong(2003)
points out，this last implication is in sharp contrast with historical experience．Several
economic histoflans have documented that the Industrial Revolution in England was
19
山东大学硕士学位论文
either preceded or accompanied by an agricultural revolution，in which the
productivity of agriculture is extremely increased．(See，e．g．，Allen，1994；Overton，
1996；and Clark，1999．)Stmctural transformations，which carl be defined嬲a secular
decline of agriculture’s share in both employment and output，in many Asian
countries also started with a period of rapid productivity growth in agriculture．Thus
to model the transformation of traditional agriculture，Xiaodong Zhu and Dennis Tao
Yang西ve another model of transition and transformation，in which the structural
transformation plays an important role．
5．2 the Model ofAgricultural Revolution
The term agricultural is not clear．Transforming traditional agriculture is more
appropriate．But in order to emphasize the importance of agriculture the term
“a母-icultural revolution”is still used here．
5．2．1 the Question ．
Sustained improvements in living standards that we see in modem industrialized
countries are a very recent phenomenon．Prior to 1820，the world economy was in a
so-r．alled Malthusian state with very little growth in output per worker．Since 1820．
the worid’s output per worker doubled itself three times．What caused the trailsition
from the Malthusian state to modem growth?Why the transition started around 1800．
not earlier?Satisfactory answers to these questions are important not iust in terms of
economic history；they may also help us to understand the income disparities we
observe today across countries．As pointed out by Lucas(1998 and 2000)，the timing
of a country’s transition directly affects the country’s position in the world income
distribution today．删le the timing ofthe transition differs from country to country．
all countries that successfully went through the transition share all important colD／non
feature：a secular decline of agriculture's share in both employment and output．Those
coLIntries that have not gone throu曲the transition are also the countries that still have
a large share of labor in agriculture．This fact suggests that structural transformadon is
an important part of the transition．
5．2．2 the Answer
As emphasized by Zhu Xiaodong and Dennis Tao Yang，if we jump out of
neoelassical economics，many scholars have examined the British economy carefully，
looking for causal factors that led to the Industrial Revolution．T瓢e factors include
institutions心orth and Wei．ngast)，culture and work ethic(Max Webber，De Vries)，
and access to the resources from Americas(Pomeranz)．Xiaodong and Dennis say all
of these factors contributed to也e growth and structural transformation in England in
the 18th century．However,it argued in this paper that none of these factors may
matter if not for another revolution in agriculture that happened during the period in
the late 19th century when Britain agriculture was trailsformed by mechanization and
new inputs produced from chemical industry．Furthermore，this transformation would
not have happened without the advancement in science and technology that made the
modernization of agriculture possible．If“the百orious revolution"，the discovery of
America，and yeoman’s revolution happened 100 years early,these eventS would not
have led to an industrial revolution in the 17th century．In this sense，as XiaodonE and
20
山东大学硕士学位论文
Dennis concludes，the Briti．sh Industrial Revolution is really lucky enough．
5．2．3 the Construction of the ModeI
The model is set up in discrete time．Ttlere is a fixed amount of land in the
economy and there are identical individuals live in the economy in period t．There are
two consumption goods，agricultural and nonagricultural．Each individual household
consumes exactly constant amount of the agricultural good and uses也e rest of the
income to finance the consumption of the nonagdcultural good．Each individual lives
for one pedod and，at the end of period，gives birth to children．The land owned by the
parent will be divided equally among the children．It is assumed that the population
growth rate is a function of per capita income．
nle nonagricultural good is produced with a linear production technology．There
also exists a modem production technelegy that USeS an intermediate input，in
addition to the traditional inputs—land and labor．The intermediate input is produced
outside of agriculture．It is assumed that each unit of intermediate input requires
constant units of nonagricultural output to produce．There is a government in this
economv'which levies fiat．rate taxes on agricultural production．The tax revenues are
used to finance public consumption of nonagricultural good．
As Xiaodong and Dennis points out，their model shows that the modem
agricultural technology wiff not be adopted if the government taxes agriculture
heavily and the cost of using intermediate input is too high．n is also showed in their
model that there are thiee important factors也at have direct effects on whether the
modem technology will be used in agriculture．First,since the modem technology
requires the use of intermediate input,the cost ofproducing this input has a negative
effect on the adoption of the modem technology．Second，high productivity in
traditional agriculture，as measured by the TFP parameter，and the land to population
ratio，also has a negative effect on using modem technology．Finally，high
productivity in the nona妒cultural sector reduces the price of the nonagricultural good
and therefore the cost of intermediate input．Consequently，it has a positive effect on
the adoption decision．
Xiaodong and Dennis call the economy a traditional economy when only the
traditional technology is used in agriculture．’111us in their model it is demonstrated
that without modem agricultural technology．a traditional economy call never have a
sustained structural trailsformation and per capita income growth，no matter how
productive the nona西cultural sector is．This is one of the key conclusions of this
model．
Zhu and Yang also try to model the timing of the transition．Their conclusion is
that in the Malthusian equilibrium，agricultural labor productivity and the percentage
of labet in the nonagricultural sector are all independent of the economy’s land
endowment and the俯levelin traditional a鲥culture．However，a higher tax rate on
agriculture implies a higher level of agricultural labor productivity and a larger
percentage of labor in the nonagricultural sector．
At last Zhu and Yang give the quantitative work by simulating the model．After
the calibration of the parameters，they give their findings：
r1)Agricultural Revolution without Modemization
21
山东大学硕士学位论文
Consider an economy that is identical to the benchmark one except that the
modem agricultural technology is not available．In this case，the agricultural
revolution occurred sometime before 1800 would not have resulted in persistent
growth and structural transformation in the long run because population growth would
eventually offset alI the gains in agricultural productivity．
佗)Industrial Revolution without the First Agricultural Revolution
Suppose that English agriculture never experienced productivity improvements
during the period 150阻1800．Would there still be all industdal revolution and
structural transformation?Their figure shows that this is not an important factor in the
long run．
(3)A舒cultural Revolution in an Earlier Period
They also examize the experiment in which山e first agricultural revolution started
sometime in 1400 rather than 1500．Really as pointed in their paper，En酉and is luck．
Other nations like Netherland missed the industrial revolution due to untransformed
agricultural．IN a word timing is important，which is one of the key points of this
Paper·
5．2．4 Remarks
ne strength of this paper ties in its emphasis on the important role played by
agriculture in the transition．DHe to the interaction between industrial$octor and
agricultural sector，the transition flom Malthusian state to modem growth is inevitable．
Tllle simulated test shows tliat timing is important for industrial revolution．
This paper also contributes to policy guidance．Really as Huang Jikun said,the
obstacles that hold Chinese agriculture stagnant are mainly too causes"the fast is that
the cost in intermediate goods，Like fertilizer and tractor,is too high，and the second is
that market price of agricultural goods is distorted．The importance of these two
factors has been shown in this papen Really according to Alien Young’s explanation
of China’s ec,onomic wonder’he thinks it is not total factor pmductivity growth，but
the force of transforming from agriculture to industry that drives China．Maybe this is
not completely fight，but it does point out the important task that faces Chinese people
improve the productivity of agriculture and transfer the redundant labour force to
industry．
6．Conclusion
A theory will show more powet if it can be unified with history．Modem
economics surpasses classical economics jn its powerful analytical tools and affluent
data，and the methotis to process data．Thus as part of empiricism of economics，
economists try to unifv both modem and classical tlleories and explain both today and
history in a unified framework．neV give diffe：rent models．or theories．And each of
them emphasizes one aspect．which is important for economic growth．
Roughly we can divide all the models into endogenous and exogenous ones by
judging whether the engine of growth is determined in the model or just given out as
part of the setting of the model．Human capital．or ideas backed bv R＆D activities can
be at the choice of individual and also the engine of growth．
山东大学硕士学位论文
And fertility choice is also embodied in the quality—quantity choice of households．
Thus endogenous theory,especially human capital models．perfectly explains both
production and population．And鹤for the account of growth history production and
population are the most important aspects．Human capital models succeed in
accounting for this．Thou曲human capital models give good explanation of both
periods，they fail to endogenize the force of transition．Tllis is the blunder that human
capital models make．So wc have reasons to seek other theories．
Idea based models embody也e transition by introducing two kinds of shocks：
adoption of property rights and temporary decline in technology．And the simulated
model goes well with history data．But the key assumption that more people more
ideas is questioning．In a word，though tllis hnd model fit t11e data，its basic
assumption seems too simple and unrealistic．The salne drawbacks also lie in other
endogenous model．Really as Hansen and Prescott say endogenous models do not
have popular acceptance．Thus we need a complete new viewpoint．
Here come exogenous models．Many ideas are invented by genius and it is
reasonable to model tech progress as exogenous．Moreovcr our concerns lie in
economic growth，not in why tech progresses，and we can leave the task of
endogenizing the progress of tech to future．But there are also other drawbacks in
exogenous models．Fertility choice is not endogenized．Exogenous models seem dim
when compare,d with endogenous models in modelling fertility choice．
In a word all models are only part of the story behind the growth．Each of them
captures onlY a beautiful portrait of the real world．Every story is rigIlt and reveals
only part of the story．Thus if we collect all the stories to鼬tller'a tomplete and vivid
picture of real world，past and today,is presented before us．We need diversified
theories to explain this diversified world．None is completely fight and none is
completely wrong．Every model has its own assumptions and if only the assumptions
are satisfied the model will work．But this world is so complex that an assumption
may hold here and not there，today and not yesterday,etc．Thus all the theoretical
models arc right．under its simplified assumption of the real world．AIl the models this
paper exposes hi曲li曲t important aspects of human history of growth．And they are
not controvertible with each other．If we want to compare them and see which one is
the best，I think，the answer should be dependent on the practical problem that we
want to address．The practical problem has its specific context．or surroundings．If this
surrounding fits well with the assumption of a certain model，then we call use this
model to analyse the problem and give policy guidance．which I think is the ultimate
goal of all scientific activities．The following section 7 does this job．nIou吐“the
usefulness of useless knowledge”may digress from this paper,I feel here it penetrates
to the core of aims of research activities．
AII the models this paper exposes are neoclassical and they share some common
drawbacks．As pointed out in section 3 of this paper all models are in neoelassical
frameworks．neir attention is only around the equilibrium conditions．and about the
existence，stability,and uniqueness of the equilibrium there is no exploration into
them．The world economy today is integrated and all the models are close economies．
This also reduces the power of these models．But concerning worldwide trade is only
山东大学硕士学位论文
a recent phenomenon this assumption may hold。
The deeper limitation lies in the inability of modem economic analytical tools．
Modcm economics，especially macroeconomics，seems weak once it goes out of the
paradigm of Walrasian equilibrium．T_llough there are numerous models trying to
model non—Wjlrasian equilibrium，none gain unanimous acceptance．Classical
economists talked about behaviour of marl and the customs and institutions behind it．
Their articles are so eloquent，absorbing and enlightening that it presents a task of
modelling customs and institutions before ambitious modem economists．
Maybe someone attacks so many theories as a blind who is feeling the elephant，
we should be confident抽that our methods are scientific and we can integrate both
data and theories．Might we are shortsighted，but we are progressing and our steps are
firm，and someday we can have a clear picture of the elephant．
By now we mean economic growth，actually we talk about goods and service：
wealth only．It is worth complementing another point here：we should also care for
economic welfare．The environment and personal health are also important．And
average variable neglects one important issue in classical economics：distribution．All
these Ieave blank for us to fill in the future．
7．Poliey Guidance
Why industrial revolution happened first in England in the early 19。“Century
rather than contemporaneously or earlier in China,where the stock of usable
knowledge may have actuatly been higher?
As a Chinese one cannot help thinking of this．Thou曲China has a long history,a
great culture and a large population，it failed to escape the invasion by industrial
couutries，like many other ancient nations．wc can’t neglect the stodes told bv aU the
models也is papc『exposes．Which theory fit出e circumstances恤Cllina?Or what
conclusion can we draw from the above models?I think the answer is：human capital，
property rights and a鲥culture．
We should invest more in human capital．Generally there are two kinds of human
capital．which needs special efforts to cultivate like schooling and another kind of
human capital is gained though practice like learning by doing．Of course college
education is important,we can’t neglect other forlns of education．Vocational
education should be developed urgently in China．Those who call go to college are
only small part of young people，and workers，especially skilled workers ale great
need of today．0n{ob training is also important．Government should make it an
institution for every enterprise to have on job training．Really related policy guidance
is much．Here 1 won’t make a fuss to redo it．
Property rights can simulate people to work more efficiently．As the great idea of
“three representative”has been added into the constitution．our country is advancing
in the direction of protecting private wealth．which is accumulated by lawful methods．
E丘iciant agriculture not onlY provides more food．but also set free labor force
from lands．As Huang Jlkun said，an offieial fi'om department of agriculture．the two
山东大学硕士学位论文
factors holding Chinese agriculture from taking off like its economy are hi曲COSTS in
intermediate good and market price distortion．In brief we should cut down the price
of fertilizer and tractor’and let the price of crops to fluctuate．Most of the population
in Qina lives in mral areas and Inost of them live on agriculture．We call’t live a
prosperous life without peasants living a prosperous life．We can’t accomplish
modemization without the modernization of agriculture．Without the improvement of
productivity in agriculture the industrialization in other industrial sectors will be
postponed．In a word agriculture is the no．1 economic issue in China,and this finding
is the no．1 issue of this pap6r．
山东大学硕士学位论文
Refe-fences
I．’Becker，G⋯S Glaeser，E．L and Mushy，＆M．(1999)．Population
and economic growth．American Economic Review,Papers and
Proceedings，May，89(2)，145·149．
2．Galor，Oded and Weil，David(1999)．From Malthusian stagnation to
modem growth．American Economic Review,Papers and
Proceedings，May，89(2)，150—154．
3．Galor,Oded and w÷il，David(2000)．“Population，technology and growth：from
Malthusian stagnation tO the demographic transition and beyond．“American
Economic RevieW，September,90(4)，806—828．．
4．Goodfriend'Marvin and McDermott，John(1995)．Early development．American
Economic Review，March，85(1)，116．133．
5，HANDB00KS OF MACROECONOMICS．VOLUME ONE
6．Hansen，Gary D．and Prescott，Edward C．(2002)．MaltIlUS to Solow．American
Economic Review，September,92(4)，1205—1217．
7．Jones，Charles I．(2001)． “W缸an Industrial Revolution inevitable?Economic
growth over the very long ILia．”Advances in Macroeconomics，1(2)．
8，Kremer，Michael． “Population Growth and Technological Change：One Million
B．C．tO 1990．”Quarter扣Journal ofEconomics，August 1993，j08(3)，PP．
681—716．
9．Lee，Ronald D．，“A Historical Perspective on Economic Aspects of the
Population Explosion：The Case of Preindustrial England，”in Richard A．
Easterlin，
10．Lucas，Robert E．，Jr．‘"On theMechanics of Economic Development．’’Journal of
Monetary Economics，July 1988，22(1)，PP．3-42．
11．Lucas，R，E．Jr．“Some Macroeconomics for the 21st Century，”Journal of
Economic Perspectives，14(1)，2000，PP．159—68．
12．Lucas．R．E．Jr． “The Industrial Revolution：Past and Future．’’ in Lectures on
Economic Growth，Cambridge：Havard University Press(2002)
13．Lucas，R．E．Jr．“The Industrial Revolution：Past and Future．”in Lectures on
Economic Growth，Cambridge：Harvard University Press(2002)
14．Dennis Tao Yang and Xiaodong Zhu “Transformation of Traditional Ag订culture
and Long Term Growth”unpublished manuscript，University of Toronto(2003)
15．刘易斯著梁小民译《经济增长理论》上海三联书店上海人民出版
社l 9 9 4
山东大学硕士学位论文
致谢
感谢我的导师黄少安老师三年来对我的教导、支持与关心，没有他就不会有
我的今天获得的学术知识。桃李不言，下自成蹊，我一定要报答黄老师的。
同时也感谢王风荣老师、李保明老师、唐绍欣老师、孙天琦老师，是他们的
谆谆教诲使我进步。也感谢石莹老师在日常生活中的关心与照顾，因为我给她带
来许多麻烦。感谢中心的其他11名兄弟姐姐，感谢他们和我共享的快乐。
感谢邹恒甫老师对我的鼓励与支持。
感谢伟大的母亲，我的妈妈。
峰多巧障目，江远欲浮天。我在三年的硕士生活中最大的遗憾是没有发表满
意的文章，我会用高品质的学术论文补偿这一缺憾的。
山东大学硕士学位论文
攻读硕士期间发表的学术论文：
“国有银行上市浅析”，《西接金融》，2004年第5期，独立作者。
絮