The Depressing Effect of Agricultural Institutions on the Prewar Japanese Economy

T D E  A I
  P J E
by
Fumio Hayashi and Edward C. Prescott
February 2006
Abstract
The question we address in this paper is why the Japanese miracle didn’t take place
until after World War II. For much of the pre-WWII period, Japan’s real GNP per worker
was not much more than a third of that of the U.S., with falling capital intensity. We
argue that its major cause is a barrier that kept agricultural employment constant at about
14 million throughout the prewar period. In our two-sector neoclassical growth model,
the barrier-induced sectoral mis-allocation of labor and a resulting disincentive for capital
accumulation account well for the depressed output level. Were it not for the barrier,
Japan’s prewar GNP per worker would have been close to a half of the U.S. The labor
barrier existed because, we argue, the prewar patriarchy, armed with paternalistic clauses
in the prewar Civil Code, forced the son designated as heir to stay in agriculture.
Keywords: prewar Japan, agriculture, barrier to labor mobility, two-sector growth model.
Earlier versions of the paper were presented on various occasions, including the 2003 Australasian Econometric
Society Meeting in Sydney, the 2004 NBER Japan program meeting in Tokyo, the 2004 Taipei Conference on
Macroeconomics and Development, and workshops at Arizona State University and University of Pennsylvania.
We thank Serguey Braguinsky, Toni Braun, Steve Davis, John Fernald, Richard Rogerson, and David Weinstein
for comments on earlier versions of the paper. Shuhei Aoki provided competent research assistance. Gratefully
acknowledged is research support from Grants-in-Aid for Scientific Research No. 12124202 administered by the
Ministry of Education, Culture, Sports, Science and Technology of the Japanese government and the National Science
Foundation SES 0422539.
1

---

1.
Introduction
The Japanese miracle, which lifted the Japanese economy from the ashes of the World War II destruction
to the present-day prosperity, and the “lost decade” of the 1990s, during which the economy ceased to
grow, are well known. Much less so is the decades-long stagnation before World War II: for much of the
prewar period of 1885-1940, Japan’s real GNP per worker remained a third of that of the U.S. This paper
addresses the question of why the Japanese miracle didn’t occur before the war.
An amazing fact about agricultural employment in prewar Japan is that it was virtually constant
at 14 million persons (about 64% of total employment in 1885) throughout the entire prewar period,
despite a very large urban-rural income disparity. There was a persistent and significant rural-to-urban
emigration, but its size was never large enough to diminish rural population. The constancy strongly
suggests that there was a powerful non-economic force that prevented people from moving out of
agriculture and that the 14 million was a binding lower bound for agricultural employment.
This leads us to ask whether a possible sectoral mis-allocation of labor due to this barrier to labor
mobility had a quantitatively important effect on the economic development of prewar Japan. We do so
by a two-sector neoclassical growth model with agriculture. Our two-sector model builds on the long
tradition of modeling the “dual economy” starting with Jorgenson (1961).1 Its most recent renditions
are Echeverria (1997), Robertson (1999), Laitner (2000), Gollin, Parente, and Rogerson (2002), and others.
They examine the interaction of Engel’s Law and the dynamics of capital accumulation, which also is a
focus of our paper.
Our paper is also related to the recent development accounting literature on international income
differences, although our focus is exclusively on Japan in relation to the U.S.2 Vollrath (2005) finds for a
number of countries that the sectoral allocation of labor and capital is inefficient because their marginal
products are not equated across sectors. He finds that the factor market inefficiency accounts 90 to
100% of international differences in the overall (economy-wide) TFP. Restuccia, Yang, and Zhu (2004)
show that sectoral distortions in the use of intermediate inputs and labor generate large international
differences in the overall TFP. Our paper is similar to these papers in that we find a large effect on the
overall TFP of a sectoral distortion (due, in our case, to the labor barrier) for prewar Japan. Unlike these
1Thus, contrary to the Lewisian tradition of development economics, we that agricultural workers are paid their
marginal, not their average, product.
2See Caselli (2004) for a survey of the recent burgeoning literature on development accounting.
2

---

papers, however, our analysis is dynamic in that we also examine the effect of the distortion on capital
accumulation.
We show that our simple two-sector growth model, when the labor barrier is superimposed on it, can
account for the prewar Japanese stagnation. We then lift the labor barrier to predict what would have
happened to the prewar Japanese economy. This counter-factual simulation shows that prewar GNP per
worker would have been higher by about 33%, which would have placed Japan at close to a half, not
a third, of the U.S. in terms of per-worker output. This substantial output gain comes about despite a
tight grip of Engel’s law. In our model, minimum food consumption is set at a very high fraction, 90%,
of food consumption in 1885. Given the state of technology for food and non-food production, however,
the country did not need 64% of total employment in agriculture to feed its population in 1885 and after.
The labor force released from agriculture upon the removal of the labor barrier would have been put to
better use outside agriculture and the increased overall production efficiency would have sparked an
investment boom.
The plan of the paper is as follows. The next section, Section 2, describes facts about Japan’s economic
development in more detail. Section 3 advances our sectoral mis-allocation hypothesis and summarizes
our main results. This is followed by four sections of elaboration: a presentation of the two-sector growth
model in Section 4, asymptotic properties of the model in Section 5, a calibration of the model in Section
6, and simulation results in Section 7. The issue of why there existed the barrier in the first place will be
taken up in Section 8. Section 9 briefly states conclusions and an agenda for future research.
2.
Accounting for Japan’s Economic Development Since 1885
The Postwar Miracle and Prewar Stagnation
We start out with a look at aggregate output since 1885. The basic data source for prewar macro variables
for Japan (Japan proper, excluding former colonies) is the Long Term Economic Statistics (hereafter LTES),
a consistent system of national income accounts compiled by a group of Japanese academics. The reader
is referred to Appendix 1 for how real GNP and other variables are constructed from the LTES and other
sources. Figure 1 shows output per worker (real GNP divided by working-age population) for Japan
3

---

and the U.S for 1885-2000, with Japan assumed to be about 71% of the U.S. for 1990.3 The scale is in logs
with a base of 2, so a difference of 1.0 in the log scale represents a 100% difference in levels. The log
difference for 1885 is 1.75, which means U.S. per-worker GNP was about 3.4 (= 21.75) times Japan’s (or
Japan was about 30% of the U.S.) in 1885.
Figure 1: GNP Per Worker, 1885-2000
¥
¤
£
&
%
"
$
#
"
!




¢

¡
 
¦
§
§
¦
§
©

¦
§
©
¦
©


¦
©

¦
©
¦

¦
©
¦
¦
©


¦
©

¦
©


¦
©

¦
©


¦
©

¦
©

¦
©
¦
©


¦
©

¦
©

¦
©
¦
©
§

¦
©
§
¦
©
©

¦
©
©




¨
¨
¨
¨
¨
¨
¨
¨
¨
¨
¨


¨
¨
¨
)
'
1
2
1
3
(
(
0
There are three features that would catch anyone’s eye. The first is the fabulous growth in the post-
World War II era, known as the Japanese miracle. There was a five-fold increase in Japan’s GNP per
worker in 25 years since 1947. The second is the prewar stagnation of several decades: between 1885
and 1940, Japan’s GNP per worker remained 30% to 50% of the U.S. GNP per worker. The average of
the Japan-U.S. ratio of per-worker output is 36%. The third feature is the stagnation in the 1990s. We
have dealt with Japan’s 1990s elsewhere (Hayashi and Prescott (2002)). The question we address in this
paper is why the Japanese miracle didn’t take place until after World War II.
3U.S. real GNP since 1929 is the chained real GNP from the U.S. NIPA (National Income and Product Accounts).
Real GNP before 1929 is from Balke and Gordon (1989). The working-age population is the population 16 years or
older for the U.S. and 15 or older for Japan until 1945 and 21 years or older for the U.S. and 20 or older for Japan after
1945. We assume that the ratio of Japan’s per-worker GNP to that of the U.S. is 71.2% in 1998. This ratio is implied
by Maddison’s (2001) estimate of GDP levels in 1990 international PPP dollars, placing Japan’s GDP at 31.9% of U.S.
GDP for 1990. See his Tables A1-b and A-j.
4

---

Growth Accounting
A very standard way to account for a country’s growth is to define the (overall or macro) TFP (total
factor productivity) as
Y
TFP
t
t ≡
,
(2.1)
Kθ (h
t
t Et)1−θ
where Yt is aggregate output in period t, Kt is aggregate capital stock, Et is employment, ht is average
hours worked per employed person (so htEt equals total hours worked), and θ is capital’s share of
aggregate income. In the growth accounting calculations below, we set θ to the customary value of 1/3.
An easy algebra on this definition yields that GNP per worker can be decomposed into four factors:
θ
Yt
1
K 1−θ
E
= TFP 1−θ ×
t
×
t
× h
N
t
t,
(2.2)
t
Yt
Nt
where Nt is working-age population. This formula shows that, in the long run where the capital intensity
(i.e., the capital-output ratio, Kt ), the employment rate ( Et ), and hours worked per employed person
Yt
Nt
1
(h
1−θ
t) are constant, the trend in GNP per worker ( Yt ) is given by the TFP factor TFP
. The power 1
Nt
t
1−θ
represents the magnification effect of TFP that an increase in TFP generates a proportionate increase in
θ
the capital stock, so the capital intensity factor ( Kt 1−θ ) represents only the part of capital accumulation
Yt
not induced by TFP growth.4 The left-side, GNP per worker, has been graphed in Figure 1.
Table 1 reports the average annual growth rate of per-worker GNP and its four factors shown in
(2.2) for prewar and postwar Japan.5 For the high-growth era of 1960-73, despite a decline in both the
average hours worked per worker and the employment rate, a high per-worker GNP growth rate of
7.2% is brought about by a very high TFP growth and, less importantly, by a slight increase in capital
θ
intensity (a 0.8% growth in Kt 1−θ ). For the prewar period, there was no increase in capital intensity:
Yt
between 1885 and 1940, the capital-output ratio declined. The much lower TFP growth before the war
than after means that the prewar TFP level was very low. This is described by the graph of the TFP factor
1
TFP 1−θ in Figure 2 (for now, ignore the line labeled ”without barrier”). Postwar TFP factor lies far above
t
the dotted trend line extrapolated from the prewar period.
4This formula has been adopted by King and Levine (1994), Hayashi and Prescott (2002) and others. Klenow and
Rodriguez-Clare (1997) has a discussion of the advantages and disadvantages of this form of growth accounting
against the more standard form of decomposing output growth into TFP, capital growth, and labor growth.
5We use the deflator for non-agricultural goods to convert nominal capital stock into real capital stock in
calculating K. This is to be consistent with the assumption of the paper’s model that agricultural goods cannot be
used as an investment good. See Appendix 1 for more details on how we defined real output Y and the real capital
stock K. Employment and hours worked are not adjusted for quality. The initial year for the postwar period is taken
to be 1960 because the capital stock data in the Japanese national accounts for the early 1950s seems unreliable.
5

---

Table 1: Growth Accounting
average annual growth rate (in percents) of
per-worker GNP
TFP factor
capital intensity
employment
hours worked
Y
1
t
θ
1−θ
rate Et
per worker h
N
TFP
1−θ
t
t
t
factor Kt
Nt
Yt
1885-1940
2.1%
2.9%
-0.6%
-0.4%
0.2%
1960-1973
7.2%
7.3%
0.8%
-0.7%
-0.2%
Note: Geometric means. Yt = GNP, Kt = capital stock, Et = employment, Nt = working-age
population, ht = average hours per employed person. See (2.1) for the definition of TFPt. θ = 1/3.
Figure 2: Japan’s Overall TFP Factor, 1885-2000
@
9
8
a
`
W
Y
X
7
W
V
U
T
S
R
Q
6
5
4
A
B
B
A
B
D
E
A
B
D
A
D
E
E
A
D
E
A
D
A
E
A
D
A
A
D
F
E
A
D
F
A
D
G
E
A
D
G
A
D
H
E
A
D
H
A
D
E
A
D
A
D
I
E
A
D
I
A
D
E
A
D
A
D
B
E
A
D
B
A
D
D
E
A
D
D
F
E
E
E
C
C
C
C
C
C
C
C
C
C
C
P
P
C
C
C
b
b
h
q
r
q
b
d
d
d
t
d
c
c
e
f
g
f
s
c
u
u
g
u
v
u
g
c
u
u
g
w
ip
p
To summarize, Japan’s prewar stagnation can be accounted for by the low level of overall TFP and
falling capital intensity.
6

---

3.
The Basic Idea and Summary of Results
The Sectoral Mis-allocation Hypothesis
The thesis of this paper is that the labor barrier — a barrier limiting the extent of rural-to-urban emigration
— contributed to Japan’s pre-World War II stagnation. We were led to this thesis by the following
observations. Figure 3 shows that employment in agriculture (here and elsewhere excluding forestry
and fishery) was essentially constant at 14 million persons in prewar Japan.6 This figure strongly suggests
that agricultural employment was constrained to be at least about 14 million. The figure also shows that,
in sharp contrast to the prewar era, postwar Japan witnessed a steep decline in agricultural employment.
As the labor force expands, a constant level of employment means a slowly declining employment share.
This is shown in Figure 4, where the employment share of agriculture declined only gradually before
the war and very sharply postwar. It took Japan 50 years to reduce the agricultural employment share
from 60% to 40%. In most other developed countries, the decline was faster: of 16 developed countries
examined in Maddison (1991, Table C.5), agriculture’s employment share in 1950 is highest for Japan
(48.3%), followed by Finland (46.0%) and Italy (45.4%).
6The source is Table 33, Volume 9 (covering agriculture) of LTES published in 1966. The estimation of employment
in the LTES was carried out under the leadership of Mataji Umemura. His 1968 estimate of employment in
agriculture (Umemura (1968)) differs from his 1966 estimate in that agricultural employment slowly declined to
about 16 million by 1914, declined sharply to 14 million between 1914 and 1919, and stayed more or less constant at
14 million thereafter. However, more recent LTES estimate, included in Volume 2 (the lead author of the volume is
Mataji Umemura) published in 1988, has agricultural employment virtually identical to the Volume 9 estimate. See
Appendix 1 for a summary of the methodology used in estimating agricultural employment in the LTES.
7

---

Figure 3: Employment in Agriculture, 1885-2003
ƒ
‚
ƒ

ƒ
€
ƒ
y
f
˜
—
ƒ
x
f
e
d
™
˜
—
‚
•
–
–
•
”

€
y
x
„
…
…
„
…
‡
ˆ
„
…
‡
„
‡
ˆ
ˆ
„
‡
ˆ
„
‡
„
ˆ
„
‡
„
„
‡
‰
ˆ
„
‡
‰
„
‡

ˆ
„
‡

„
‡
‘
ˆ
„
‡
‘
„
‡
ˆ
„
‡
„
‡
’
ˆ
„
‡
’
„
‡
ˆ
„
‡
„
‡
…
ˆ
„
‡
…
„
‡
‡
ˆ
„
‡
‡
‰
ˆ
ˆ
ˆ
†
†
†
†
†
†
†
†
†
†
†
“
“
†
†
†
Figure 4: Agriculture’s Employment Share
i
g
g
h
q
g
h
p
g
h
g
h
o
n
g
h
g
h
m
l
g
h
k
g
h
j
g
h
i
g
h
g
h
r
s
s
r
s
u
v
r
s
u
r
u
v
v
r
u
v
r
u
r
v
r
u
r
r
u
w
v
r
u
w
r
u
x
v
r
u
x
r
u
y
v
r
u
y
r
u
v
r
u
r
u
z
v
r
u
z
r
u
v
r
u
r
u
s
v
r
u
s
r
u
u
v
r
u
u
w
v
v
v
t
t
t
t
t
t
t
t
t
t
t
{
{
t
t
t
Thanks to the labor barrier setting the binding lower bound of 14 million on agricultural employment,
there was too much labor tied up in the decreasing-returns-to-scale technology called agriculture. For
reasons explored in Section 8, this barrier ceased to operate after World War II, which must have
contributed at least in part to the rapid increase in the overall TFP in postwar Japan. Hansen and Prescott
8

---

(2002) described the industrial revolution as a switch from a decreasing-returns-to-scale with respect to
reproducible inputs technology (the Malthus technology) to a constant-returns-to-scale technology (the
Solow technology). Our hypothesis can be rephrased as saying that the transition from Malthus to Solow
was inhibited by the barrier to labor mobility.
Main Results
The rest of this paper formalizes our sectoral mis-allocation hypothesis for prewar Japan in a two-sector
neoclassical growth model with agriculture and see to what extent the model with labor barrier can
account for the prewar stagnation. The model’s main features are as follows: (i) the actual path of
sectoral TFPs (shown in Figure 5) is taken as given and was perfectly foreseen by agents as of 1885, (ii) the
path of total employment is exogenously given to the model, (iii) the sectoral allocation of capital as well
as capital accumulation are endogenous, (iv) the sectoral allocation of employment is endogenous, albeit
subject to the labor barrier that sets the lower bound of 14 million for employment in the agricultural
sector, and (v) a stringent Engel’s law that the subsistence level of food consumption per worker is 90%
of per-worker food consumption in 1885. The paper’s main results are the following.
Figure 5: Sectoral TFPs, 1885-1940 (1885=100)

|
|
~
|
|
}
|
|
|
}
€
€
}
€
‚
|
}
€
‚
}
‚
|
|
}
‚
|
}
‚
}
|
}
‚
}
}
‚
~
|
}
‚
~
}
‚

|
}
‚

}
‚
ƒ
|






‡
Š
‡
Š
‹
‹
„
…
†
ˆ
‰
‰
†
Œ

Ž

„
…
†
ˆ
‰
‰
†
Œ

• Our simple two-sector model can account for the prewar stagnation. The simulation (i.e., the solution
path of the model) in which the labor barrier is imposed tracks the prewar data closely. This is shown
for per-worker GNP in Figure 6: the simulation represented by the dotted line labeled “with barrier”
9

---

does not depart substantially from the actual — except for the post-World War I period where the
model consistently over-predicts output (see below for more on the over-prediction for the interwar
period). The lower bound for agricultural employment is binding throughout the prewar period.
Figure 6: GNP Per Worker, 1885-1940
•

‘
”
’
‘
”

‘
¨
§
¤
¦
¥
“
’
¤
£
‘
¢
¡
 
Ÿ
ž
“

‘

’
‘


‘
–
—
—
–
—
™
š
–
—
™
–
™
š
š
–
™
š
–
™
–
š
–
™
–
–
™
›
š
–
™
›
–
™
œ
š
–
™
œ
–
™

š
˜
˜
˜
˜
˜
˜
©
«
³
µ
·
¹
º
·
³
µ
·
¹
º
·
¸
¸
¸
­
®
­
¯
­
®
­
¯
²
´
­
»
»
´
»
­
®
­
¯
²
´
²
¼
­
»
»
´
»
ª
ª
¬
¬
°
±
° ¶
¬
°
±
°¶
• The quantitative effect of the labor barrier is large. The counter-factual simulation — the solution path
of the model that does not impose the labor barrier but still takes the sectoral TFPs as given — for
per-worker GNP is the line labeled “without barrier” in Figure 6. Japan’s GNP per worker, which was
about a third of the U.S. level for much of the prewar period, would have been substantially higher
(see Table 2 below for more quantitative information), were it not for the barrier.
• There are two main sources of this big gain in aggregate output. The first is a production efficiency
gain due to the removal of the barrier. Going back to Figure 2, also shown in the figure is the overall
TFP factor calculated from the counter-factual simulation. It shows that the removal of the barrier-
induced mis-allocation of labor significantly raises the overall TFP. (The efficiency gain declines with
time because the lower bound for agricultural employment gradually falls as a fraction of total
employment.) The second is an investment boom sparked by the increased production efficiency.
Figure 7 plots the capital stock per worker, Kt/Nt, for the three cases, actual and simulations with
and without the barrier. In the counter-factual simulation (without the barrier), the capital stock rises
10

---