Upload
buidat
View
216
Download
2
Embed Size (px)
Citation preview
END
NEO
SUBDOGEN
O-SCH
U
Na
BSIDIENOUS
IHUMPE
Univers
Depaationa
DISC
ECO
ES IN CYCLNVES
ETERIA
Businsity of
Jie
artmenal Univ
CUSSIO
ONOM
AN ECLES OVTMENAN INN
by
Bei Li
ness Sf West
and
e Zhan
nt of E
versity
ON PA
MICS
CONOVER N
NT ANDNOVA
i
Schooltern Au
ng
Econoy of Sin
APER
OMY WNEOCLD
ATION
l ustral
micsngapo
11.23
WITH LASSI
REGIM
ia
ore
CAL
MES
1
SUBSIDIES IN AN ECONOMY WITH ENDOGENOUS CYCLES OVER
NEOCLASSICAL INVESTMENT AND NEO-SCHUMPETERIAN INNOVATION REGIMES *
Bei Li
Business School University of Western Australia
and
Jie Zhang
Department of Economics
National University of Singapore
July 5, 2011
DISCUSSION PAPER 11.23 ABSTRACT
We explore the roles of subsidies in the Matsuyama model (1999) of growth through cycles alternating perpetually between two phases featuring neoclassical investment and neo-Schumpeterian innovation respectively. Subsidies to R&D investment or to the purchase of newly invented intermediate goods can arbitrarily reduce the threshold level of capital per type of intermediate good, beyond which the economy moves from the investment phase to the innovation phase. More importantly, such subsidies can mitigate and eventually eliminate cycles for significant welfare gains that can be equivalent to as much as 10% rises in consumption at all times. Keywords: Subsidization; Innovation; Capital accumulation; Cycles; Growth JEL Classifications: E3, E6, H2, H3, O4 Correspondence: Professor Jie Zhang, Department of Economics, National University of Singapore, Singapore 117570; [email protected]; Tel: +65 6516 6024; Fax: +65 6775 2646 *We would like to thank the participants at the 10th SAET Conference (Society for the Advancement of Economic Theory, 2010) for their valuable comments. All errors or omissions are our own.
2
1. Introduction
One theme of the macroeconomic aspect of public policy analysis is to explore whether and
how government policies can promote output growth or mitigate output fluctuations for
welfare gains. In this regard, different macroeconomic models have different answers. The
neoclassical growth model, pioneered by Solow (1956) and Swan (1956), captures how
capital accumulation contributes to growth and why growth eventually halts at a stable
steady-state output level per capita under diminishing returns to investment. It may need no
role of government intervention so long as consumers choose their consumption path
optimally as in Cass (1965) and Koopmans (1965) according to the Welfare Theorems. This
policy implication remains valid even when exogenous shocks are introduced into the
neoclassical growth model for the creation of cycles as in the real business cycle models led
by Kydland and Prescott (1982) and Long and Plosser (1983).
The neo-Schumpeterian models developed in the last two decades generate sustainable
growth through costly innovations that create new varieties of intermediate goods or
improve the quality of existing intermediate goods (see, e.g. Romer, 1990; Aghion and
Howitt, 1992). In these R&D endogenous growth models, R&D activities intended for a
new variety or a quality improvement incur a fixed cost, whereas the production of each
intermediate good incurs a constant marginal cost; and both the innovation and the
intermediate goods production use current final output. Monopoly rights are granted to
innovators in order to allow them to recoup their innovational costs. The consequent
monopoly pricing reduces the demand for new intermediate goods, causing lower final
output and slower growth than their socially optimal levels (static and dynamic losses in
3
efficiency, respectively). The efficiency losses of monopoly pricing justify government
subsidization either to R&D investment or to the purchase of newly created intermediate
goods, as shown in Barro and Sala-i-Martin (1995) with inelastic labor, and in Zeng and
Zhang (2007) with elastic labor, among others. In contrast to the neoclassical growth and
the real business cycle models, however, the economy is always on the balanced growth
path in such neo-Schumpeterian growth models that are known as the AK model in essence.
Matsuyama (1996, 1999) unifies the neoclassical and the neo-Schumpeterian growth
models by assuming that both R&D activities and intermediate goods production can only
use available capital from previous savings, among various approaches generating
sustainable growth through endogenous cycles.1 Under this neoclassical-style assumption,
innovation can break even to recover the fixed cost when capital per type of intermediate
good exceeds a critical level for a profitable scale of the demand for newly invented
intermediates. Once currently available capital exceeds this level and induces innovation,
however, part of the capital stock must be used for the fixed innovation cost and the
remaining amount of capital for manufacturing intermediate goods declines. Consequently,
current innovation, if responding more elastically than capital investment to initial
abundance in capital per variety, can reduce capital per variety to the extent such that future
innovation becomes unprofitable until enough capital is formed again through a neoclassical
1 Freeman, Hong and Peled (1999), Maliar and Maliar (2004), and Walde (2002, 2005) also attain sustainable growth through endogenous cycles in models with capital accumulation and innovation. In their models, innovations advance non-rivaled general technology for final goods production, rather than rivaled intermediate goods sold under exclusive monopoly rights in the R&D growth models. Francois and Lloyd-Ellis (2003) acknowledge the relevance of such major breakthroughs in general technology in explaining "long waves" of fluctuations but cast doubt on the relevance in explaining high-frequency business cycles. To focus on high-frequency cycles, they assume multi-sectors and argue for endogenous clustering of innovation and cycle implementation. In Bental and Peled (1996), firms producing final goods engage in costly search of better technology from a known and fixed pool of technologies. We follow the approach of Matsuyama (1999) assuming that costly innovations lead to rivaled intermediate goods sold monopolistically, because monopoly pricing creates inefficiency and thus calls for government intervention.
4
investment phase. It argues that this is an empirically plausible scenario: The balanced
growth path with innovation is unstable and therefore the economy may fluctuate between a
Solow investment phase and a Romer innovation phase perpetually, causing cyclical
movements in consumption, investment, innovation and growth.
In a different type of neo-Schumpeterian growth model by Francois and Shi (1999) and
Haruyama (2009), labor rather than capital is the sole input for innovations and for
intermediate goods production.2 In their models, cycles can also arise endogenously from
contemporaneous complementarities between investors devoting labor to innovation for
temporary profits. Haruyama (2009) demonstrates that optimal steady-state R&D
subsidization fails to eliminate cycles and should be state-dependent in a fluctuating
economy when labor is the only input for innovation and intermediate goods production.
The cyclical fluctuations in consumption, investment, innovation and growth represent
another source of efficiency loss, given diminishing marginal utility of consumption and
diminishing marginal product of factors, particularly when innovation and intermediate
goods production compete with each other for limited available capital. This adds to the
efficiency losses of monopoly pricing analyzed in the literature and thus differs from the
Pareto optimal variations in consumption across time and sectors facing exogenous shocks
found in Long and Plosser (1983). Therefore, important macroeconomic questions arise as
follows. Can government policies mitigate or eliminate such cyclical fluctuations and
promote innovation, investment and growth at the same time when innovation and
intermediate goods production require capital accumulated from previous savings? Can
2 This assumption is made in one line of the R&D literature, e.g. Grossman and Helpman (1991), Segerstrom (1991), Aghion and Howitt (1992), Young (1993), Parente (1994), and Fan (1995).
5
such government policies enhance social welfare?
When attempting to answer these questions, it is natural to focus on subsidization
associated with R&D activities or with the purchase of new R&D products. Although such
subsidization has been considered in the literature with innovation and growth as mentioned
earlier, the existing studies have mainly focused on how the subsidies mitigates the losses in
efficiency of monopoly pricing on a stable balanced growth path.
To our knowledge, the only work on subsidization in the unified model of Matsuyama
(1999), where innovation and intermediate goods production cost available capital, is Aloi
and Lasselle (2007). They use a lump-sum subsidy to innovators financed by a lump-sum
tax on consumers and find that it can promote growth, stabilize the innovation cycles and
increase welfare. However, their subsidy adds directly to capital in the resource constraint
for innovation and intermediate goods production in the current period. In the Matsuyama
model, by contrast, the constraint on available capital from previous saving for current
innovation and intermediate goods production is essential for the creation of cycles. It thus
remains to show whether and how more realistic subsidization, which does not relax this
assumption, can induce changes in individuals' behavior so as to stabilize the balanced
growth path for welfare gains.
In this paper, we explore whether flat-rate subsidization associated with R&D activities
can mitigate or eliminate cycles for welfare gains in the Matsuyama model. In doing so, the
subsidy provides additional awards to innovators without relaxing the constraint on
available capital for innovation and intermediate goods production. We find that subsidies to
R&D investment or to the purchase of newly invented intermediate goods, financed by a
6
consumption tax, can arbitrarily reduce the threshold level of capital per variety, beyond
which the economy moves from the investment phase to the innovation phase. Moreover,
sufficient subsidization can change the balanced growth path from an unstable one to a
stable one and thus eventually eliminate the cycles. In numerical examples for plausible
parameterizations, optimal subsidy rates can achieve substantial welfare gains (equivalent to
as much as about a 10% increase in consumption in every period) and lie in the range that
leads to convergence toward a stable balanced growth path.
Our results appear consistent with the postwar experiences in some industrial nations
such as the United States where substantial subsidies are provided to R&D activities and to
the purchase of new equipment. For example, there is substantial subsidization in the US tax
system:a 50% immediate writing-off of equipment investment, expensing of R&D
expenditures, and accelerated depreciation allowances, according to Gordon, Kalambokidis,
Rohaly and Slemrod (2004), Gordon, Kalambokidis and Slemrod (2004), and others. At the
same time these countries observe much more innovations but dampened recessions
compared to previous times on average.
The rest of the paper proceeds as follows. Section 2 introduces the building blocks of
the model. Section 3 characterizes the steady states in different regimes and analyzes the
global dynamics for different levels of subsidization. Section 4 deals with optimal subsidy
rates and presents numerical simulation results. Section 5 concludes.
2. The model
The model is an extension of Matsuyama (1999, 2001) by considering subsidies to R&D
7
spending and to the purchase of newly invented intermediate goods, financed by a
consumption tax. Time is discrete, extending from period one to infinity � � 1,2, … ,∞.
2.1. Production and innovation
There is a single consumption-investment good taken as a numeraire and produced by using
capital and labor. Labor is supplied inelastically at an amount L that also stands for the
size of the population of identical, infinitely-lived consumers or workers. Let1tK − denote the
capital stock available for the creation and production of intermediate goods in period �. In
the first period, there is an initial capital stock 0 0K > .
Capital is first converted into a composite of intermediate goods by a symmetric CES
function. Let ( )tx z denotes the th type of available intermediate good in the range
[0, ]tN in period t . Labor and the composite of intermediates are combined through a
Cobb-Douglas technology for final goods production:
( ) ( ){ }1 11
0
tN
t tY A L x z dzσσ −
= ∫ , (1)
where 0A > is the total factor productivity parameter and 1σ > is the direct partial
elasticity of substitution between every pair of intermediate goods.
One unit of each type of intermediate good is manufactured by converting a units of
capital. In each period t, old intermediate goods in the range [ ]10, tz N −∈ starting with
0 0N > are sold competitively, while new intermediate goods in the range [ ]1,t tz N N−∈
may be introduced and sold exclusively by their innovators under a one-period patent
protection. Creating a new intermediate requires F fixed units of capital. By symmetry, we
have ( ) ct tx z x≡ for competitively supplied intermediate goods [ ]10, tz N −∈ , and
8
( ) mt tx z x≡
for monopolistically supplied intermediate goods 1,t tz N N− ∈ . The profit
function for firms in the final goods sector can be expressed as:
( ) ( ) ( )( )1 1 1 11
1 1c m
t t t t t tA L N x N N xσ σσ − −
− − Π = + −
( )( )1 11c c m mt t t x t t t t tN p x s N N p x w L− −− − − − − , 0 1xs≤ < , (2)
where ctp and m
tp are the prices of old and new intermediate goods, respectively; xs is
the time-invariant subsidy rate to the purchase of new intermediate goods used in Barro and
Sala-i-Martin (1995) and Zeng and Zhang (2007) but not considered in Aloi and Lasselle
( 2007); and tw is the wage rate per unit of labor. According to Zeng and Zhang (2007),
the subsidies to final output or to the purchase of intermediate goods are equivalent
concerning their effects on growth and welfare. Thus, we only consider the latter.
In the final goods sector, factors are paid by their marginal products:
( ) ( ) ( ) 111 1c c
t tp A L xσσσ
−= − , (3)
( ) ( ) ( ) 111 (1 1 ) ,m m
t x tp s A L xσσσ
−− = − (4)
1 tt
Yw
Lσ= . (5)
Let tr denote the price of capital. Then the marginal cost of manufacturing intermediate
goods in period t is equal to tar . All old intermediate goods are supplied competitively at a
price level equal to the marginal cost, ( ) ct t tp z p ar≡ = for [ ]10, tz N −∈ , while all new
intermediate goods, once introduced, are sold monopolistically at a higher price level,
( ) ( )[ 1 ]m ct t t tp z p ar pσ σ≡ = − > , for 1,t tz N N− ∈ , following from (4). From equations (3)
and (4), the relationship between ctx and mtx must satisfy:
( ) ( )11 1
1
c ct t
xm mt t x
x ps
x p s
σ σσ
σ
− − = = − − − . (6)
9
Absent subsidies, the higher price of new intermediates than that of old intermediates
yields a smaller equilibrium quantity of each new intermediate than that of each old
intermediate, �� �
�. This asymmetry in the equilibrium quantities of new versus old
intermediates must lead to static and dynamic efficiency losses. The lower demand for new
than for old intermediate goods leads to a dynamic efficiency loss in terms of decelerating
the rate of innovation because it reduces the profitability scale for innovators to recover the
fixed R&D cost. At the same time, the lower demand for new than for old intermediate
goods leads to a static efficiency loss in terms of decreasing final output because all
intermediates enter final goods production symmetrically and have diminishing marginal
contributions to final output.
Subsidies on the purchase of new intermediate goods strengthen final goods producers'
demand for new intermediate inputs relative to old ones by reducing the user cost of new
intermediates. According to (6), when 1/xs σ< , c mx x> ; when 1/xs σ= , c mx x= ; when
1/xs σ> , c mx x< . Thus, the subsidy on the purchase of new intermediates may affect the
dynamic system significantly in this model.
The one period monopoly enjoyed by the innovator makes it possible to recover the
fixed R&D cost. There is free entry for innovative activities. The monopoly profit is equal
to the sales revenue net of the fixed R&D cost and the variable manufacturing cost:
( )1m m mt t t t t np x r ax s Fπ = − + − . Here, ns is the time-invariant subsidy rate to the fixed
R&D cost tr F used in Barro and Sala-i-Martin (1995) and Zeng and Zhang (2007), as
opposed to the lump-sum subsidy used in Aloi and Lasselle (2007). The free entry ensures
the following in equilibrium
10
( ) ( ) ( ) ( ) ( )1 11 1 , , 1 1 0m mt n t t t n t tax s F N N ax s F N Nσ σ− − ≤ − − ≥ − − − − = . (7)
That is, when potential innovators expect the sale of a new intermediate good to be smaller
than the break-even level (i.e. ( ) ( )1 1mt nx s F aσ< − − ), there is no incentive for innovation
at all, thereby 1t tN N −= . In equilibrium with free entry, when innovation occurs (i.e.
1t tN N −> ), the innovator must just break even such that ( 1)(1 )mt nax s Fσ= − − . By
lowering the cost of innovation virtually to any level, the subsidy on the R&D cost can
reduce this break-even level of the demand for a new intermediate virtually to any level, and
may therefore have significant effects on the dynamic system of the model.
Regardless of the value of the subsidy rates, the resource constraint on the use of
available capital for intermediate goods production and innovation in period t is:
( )( )1 1 1c m
t t t t t tK N ax N N ax F− − −= + − + . (8)
Without this assumption, the economy would always be on the unique balanced growth path
as in the earlier R&D growth models. This constraint differs from the counterpart in Aloi
and Lasselle (2007) that regards the subsidy as an addition to available capital.
Substituting equations (6) and (7) into the above constraint leads to
( ) ( ) ( ){ }1
11 1 min , 1 1c m
t x t t x nax a s x F k s sσ
σ σθσσ
−
− = − − = − −
, (9)
( ) ( ) ( ) ( )1
1 1max 0, 1 111
tt t x n t
nn
KN N s s N
ss F
σ θσσ σσ σ
−− −
= + − − − − −− − , (10)
where
( )t
tt
Kk
F Nθσ≡ ,
11
1σ
θσ
− ≡ −
, [ ]1,eθ ∈ , 2.71828...e = .
Here, θ is increasing with σ . Clearly, for innovators to break even in period t , the
amount of available capital stock 1tK − must be abundant enough relative to available variety
11
1tN − . According to (9), increasing the subsidy rate on the fixed R&D cost will reduce the
demand for both new and old intermediates, while increasing the subsidy rate on purchasing
new intermediates will reduce the demand for old intermediates, given any initial state
1 1( , )t tN K− − such that 1tk − is large enough for innovators to break even. According to (10),
increasing either of the two subsidy rates will increase the rate of innovation, given any
initial state such that 1tk − is large enough for innovators to break even.
We can now rewrite equation (1) as
( ) ( ) ( ) ( )1 1 1 11
1 1c m
t t t t t tY A L N x N N xσ σσ − −
− − = + −
. (11)
Given any initial state that allows innovators to break even, the subsidies can increase final
output by promoting innovation for faster variety expansion, but reduce final output by
reducing the demand for each intermediate good. However, the subsidy on the purchase of
new intermediates can increase final output by increasing the demand for each new
intermediate unless it is too large. To detail such effects further, we rewrite equation (11) by
using equations (7), (8), (9) and (10):
( )1 1
1 11
tt t
KY A LN
a
σσ
−−
− =
, if ( ) ( )1 1 1t c x nk k s sσ
− ≤ ≡ − − ;
( ) ( )( ) ( )
( ) ( )( )
( )
1 11 1
1 11
1 1
1 1
1
1 11 , if .
1
ntt
n
n x n nt x t c
n
F sKY A L
as F
s s s F sN s k k
as
σσ
σσ
σσ σ
σ θσσ σ
−
−
−−
− −
− − = − −
+ − − + − ≥ − −
(12)
The critical value of the capital-variety ratio, ( ) ( )1 1c x nk s sσ= − − , below which there
is no innovation and hence no subsidization by construction, divides government action in
this model into policy-dormant and policy-active regions, respectively. In fact, increasing
the rate of either subsidy can reduce the threshold level of the capital-variety ratio,ck ,
12
virtually to anywhere above zero, enhancing the chance for the economy to stay in the
policy-active region with R&D activities. Thus, the subsidization may significantly change
the dynamic path of the model.
According to (12), given an initial state 1 1( , )t tN K− − such that 1t ck k− ≥ , subsidizing
either the fixed R&D cost or the purchase of new intermediates can increase final output if
the subsidy rates are sufficiently low such that their positive impact on variety expansion
dominates. The opposite occurs for further increases in the subsidy rates if the subsidy rates
are already sufficiently high such that their negative impact on the demand for intermediates
dominates. To see this clearly, we differentiate final output with respect to one subsidy rate
at a time for any initial state 1 1( , )t tN K− − such that 1t ck k− ≥ . Focusing first on how ns
affects tY at 0xs = and 1t ck k− ≥ , /t ndY ds is signed by two parts additively. One part
containing the derivative of 1 1/(1 ) / [ ( 1)]n ns sσ σ σ−− − − with respect to ns is signed by ns− ,
while the other part containing the derivative of 1 1/(1 ) / [ ( 1)]n n ns s sσ σ σ−− − − is signed by
1 (2 1/ )ns σ− − . Combining the two parts together, /t ndY ds must be positive for very
small ns but becomes negative when ns becomes larger, at least when 1/ (2 1/ )ns σ> − .
When focusing on xs at 0ns = , /t xdY ds is signed by 1 xsσ− through signing the
derivative of 1(1 )x xs sσ −− with respect to xs . Thus, final output increases with xs when
1/xs σ< under which c mx x> ; final output peaks at 1/xs σ= whereby c mx x= : any
further increase in xs leads to c mx x< and thus reduces final output. Therefore, the
effects of the subsidies on final output also alter the static efficiency of the model.
Equations (10) and (12) are simplified to
13
( ) ( ) ( )1 11
( , , ) max 1, 1 1 11
tt x n t x n
t n
Nk s s k s s
N s
σθσψσ σ− −
−
≡ = + − − − − − , (13)
( )1
1
1 11
( , , )tt x n t
t
Y Lk s s A k
K F
σσφ
θσ−
− −−
≡ =
, if 1t ck k− ≤ ; otherwise,
( ) ( ) ( )1 1 11
11
1 1( , , ) 1 1 .
11 1
n x nt x n x n
tn
s s sLk s s A s s
F ks
σ σσ
φθσ σ
σ
− −
−−
− − = + + − − −
(14)
To economize on notations, we assume1L = , 1a = , and 1F θσ= without changing
the essence of the results.
2.2. Households and government
The infinitely lived representative agent derives utility from consumption (with1L = )
according to the following preference:
( )1
lntt
t
U Cβ∞
=
=∑ , 0 1β< < , (15)
where β is the discount factor. The logarithmic utility renders tractability.
In each period t, the agent receives capital income, 1t tr K − , and earns labor income,tw L .
He consumes tC , facing a proportional consumption tax ,c tτ , and carries over tK units of
final goods to the next period. The flow budget constraint for the agent is
( )1 ,1t t t t c t tK r K w L Cτ−= + − + . (16)
Also, the consumer faces an intertemporal solvency restriction:
1
lim 0ttts s
K
r→∞=
≥∏
. (17)
Taking initial 0 0K > and the sequences ,( , , )c t t tr wτ as given, the agent chooses the
sequence ( , )t tC K to maximize utility in (15) subject to (16) and (17). The optimal
14
intertemporal condition is
( ) ( )1
, , 1 1
1
1 1t
c t t c t t
r
C C
βτ τ
+
+ +
=+ +
, (18)
and the binding solvency condition can be written as
( )1 ,
lim lim 01
tt ttt ts s c t t
K K
r Cβ
τ→∞ →∞=
= =∏ +
. (19)
Production factors are compensated competitively according to ( )1t tw L Yσ= and
( )1 1 1t t tr K Yσ− = − , which, together with (16) to (18), yield
1
1t tK Yβσ
= −
, (20)
( )
,
1 1 1
1t
tc t
YC
β στ
− − =+
. (21)
From (20) and (21), the agent carries a constant fraction of income as capital into next
period and spends the remaining fraction on consumption in the current period. Here, a
higher tax rate on consumption spending reduces consumption (for greater subsidies).
From equations (13), (14) and (20), the dynamics of the economy can be uniquely
determined by the following system of first-order difference equations in K and N:
( ) ( )1 11 1 , ,t t x n tK k s s Kβ σ φ − −= − , (22)
( ) ( ) ( )1 1 1max 0, 1 11 1 1t t t x n t
n
N N K s s Ns
σθσ− − −
= + − − − − − , (23)
starting from an initial state 0 0( , )K N , given time-invariant subsidy rates, xs and ns .
Observe in (23) that sufficient subsidization of either type can lead to the introduction of
new intermediates, 1 0t tN N −− > , for any initial state at time t , 1 1( , )t tN K− − .
The government runs a balanced budget in every period between taxes and subsidies:
( ) ( ) ( ), 1 11 mc t t n t t t x t t t tC s Fr N N s ar N N xτ σ σ− −= − + − − . (24)
15
Without innovation occurring ( 1t tN N −= ), the tax equals zero (the policy-dormant regime).
Combining (20), (21) and (24) together with (7) and (10) yields
( ) ( )
( ) ( ) ( )1
,
1 1
1max 0, .
1 11
t c n x n
c t
n t t c n x n
k k s s s
s k k k s s s
στ
σ β σ σ σσ
−
− −
− + − =
− − − − − + − −
(25)
From equations (22) and (23), the law of motion for the capital variety ratio, tk , is
governed by the following one-dimensional mapping, : R R+ +Φ → ,
( )( ) ( )
( ) ( ) ( ) ( ){ }( ) ( ) ( )
1
1 1
1 1
1 1 1
1
1
1
1 1
1 1 1 1 1 ,
1 1 1 1 1
t t
t t c
n t x n x n
t c
n t x n
k k
A k if k k
A s k s s s sif k k
s k s s
σ
σ σ
σ
β σ
β σ σ
σ θ
−
−− −
− −−
−
−
= Φ
− ≤ − − + − + − =
≥ − − + − − −
(26)
where ( ) ( )1 1c x nk s sσ= − − .
3. The steady state and global dynamic analysis
The steady state of the economy is defined as an equilibrium path on which t t tk K N=
stays constant over time for any time-invariant subsidy rates, xs and ns . According to (26),
the steady state k of the dynamic system is uniquely determined. Whether new
intermediates are introduced or not at the steady state depends on the relationship between
the steady state k and the critical value ck .
First, if *t ck k k= ≤ in a steady state, then according to (13) and (14), 1t tN N −= and
1t tK K −= . In this steady state, there is no innovation; all the intermediate goods are
competitively supplied; and the economy does not grow in the long run. From (22), on this
neoclassical stationary path, ( )* 1 1k Aσβ σ≡ − . The existence of such a stationary path
16
requires that ( ) ( )11 1 cA k
σβ σ− ≤ .
Now, suppose that **t ck k k= > holds in a steady state. From (22) and (23), the balanced
growth path satisfies the following:
( ) ( ) [ ]1 11 1
11 , , 1 1
1t t
t x n t ct t n
K Nk s s k k
K N s
θσβ φσ σ σ− −
− −
− = = = + − > − − .
In this steady state, the available capital stock of the economy is large enough relative to the
number of existing intermediates such that new intermediates are introduced and that tK
and tN share the same growth rate. The existence of such a balanced growth path requires
( ) ( )11 1 , , 1t x nk s sβ σ φ −− > .
These results concerning the steady state of the dynamic system in (26) are given below.
Proposition 1. Let ( ) ( )11 1 cG A k σβ σ≡ − , where ( ) ( )1 1c x nk s sσ= − − , with 0 1xs≤ <
and 0 1ns≤ < .
(1) If 1G ≤ , the dynamic system has a unique steady state * *( )k k= Φ where
( )1 1 ck A kσβ σ∗ = − ≤ . At this steady state, the economy has no innovation and does
not grow.
(2) If 1G > , the dynamic system has a unique steady state ** **( )k k= Φ where
( ) ( ) ( ) ( )1 1 1 21 1 1 1 1 1 2c n n ck k s A s k
σθ σ β σ θ−∗∗ = + − − + − − + ∆ >
,
( ) ( ) ( )
21 11 1 1 1 1 1n c ns k A s
σσ θ β σ − ∆ = − − − − − −
( ) ( ) ( ) ( )1 1 14 1 1 1 1 1n x n x nA s s s s s
σ σθβ σ σ− −+ − − − + − .
At this steady state, the economy grows in ( , )t tN K at the same constant rate
( ) ( ) ( )1 1 , , 1 .
1x n c
n
g k s s k ks
θσβ σ φσ σ
∗∗ ∗∗ = − = + − − −
17
Proof. The solutions for the steady state *k or **k follow the respective scenarios in (26).
What remains to show is ** ck k> for 1G > . First, note the following implication of 1G > :
( ) ( )
( )( ) ( )
1 1/ 2 1/
1
2 2 1 1/ 2
1 1/ 1/ 1
[1 (1 1 / ) 1 1 (1 ) ] 4 1 1 (1 )
(1 ) [ (1 ) / ]
[1 (1 1 / ) 1 1 (1 ) ] 2 [1 (1 1 / )
1 1 (1 ) ] 4 1 1 (1 ) (1 ) [ (1 ) /
n c n c n
x x n n
c n n c n
n c n x x n n
s k A s Ak s
s s s s
k s A s k s
A s Ak s s s s s
σ σ
σ
σ σ
σ θ β σ θβ σσ
θ σ β σ θ σβ σ θβ σ σ
− −
−
−
− − −
∆ = − − − − − − + − −
− − +
= + − − − − − − − − −
− − + − − − − + ].
Note that 1 (1 1/ ) (1 )(1 ) (1 ) / 0n x n x n ns s s s s sσ σ− − − − − = − + ≥ and that 1G > implies
( ) 1 1/1 1 (1 ) (1 )(1 )n x nA s s sσβ σ −− − > − − . We can now rewrite the expression of ∆ below:
( ) ( )( )
2 2 1 1/ 2 1/
1 1 1/
[1 (1 1 / ) 1 1 (1 ) ] 2 {2 1 1 (1 )
(1 ) [ (1 ) / ] 1 (1 1 / ) 1 1 (1 ) }
c n n c n
x x n n n n
k s A s k A s
s s s s s A s
σ σ
σ
θ σ β σ θ β σσ σ β σ
− −
− −
∆ = + − − − − − + − −
− − + − + − + − −
( )( )
2 2 1 1/ 2
1/ 1
[1 (1 1/ ) 1 1 (1 ) ]
2 [1 (1 1/ ) (1 )(1 )][2 1 1 (1 ) (1 ) 1]
c n n
c n x n n x
k s A s
k s s s A s s
σ
σ
θ σ β σθ σ β σ
−
− −
> + − − − − − +
− − − − − − − − −
( )2 2 1 1/ 2[1 (1 1 / ) 1 1 (1 ) ]
2 [1 (1 1 / ) (1 )(1 )]c n n
c n x n
k s A s
k s s s
σθ σ β σθ σ
−> + − − − − − +− − − − −
( )
( )
2 2 1 1/ 2
1 1/
[1 (1 1/ ) 1 1 (1 ) ]
2 [1 (1 1/ ) 1 1 (1 ) ]
c n n
c n n
k s A s
k s A s
σ
σ
θ σ β σθ σ β σ
−
−
> + − − − − − +
− − − − −
( ) 1 1/ 2[ 1 (1 1/ ) 1 1 (1 ) ] .c n nk s A s σθ σ β σ −= + − − − − −
So ( )** 1 1/ 1/2[ 1 (1 1/ ) 1 1 (1 ) ] /(2 ) 2 /(2 )c n n c ck k s A s k kσθ σ β σ θ θ θ−= − + − + − − + ∆ > = . The
other root, ( )** 1 1/ 1/ 2[ (1 ) (1 ) (1 1/ ) 1 1 1 (1 ) ] /(2 )x n n nk s s s A sσ σθ σ β σ θ−= − − + − − + − − − ∆ ,
is dropped for being inconsistent with ck k∗∗ > . Q.E.D.
The implication of Proposition 1 is as follows. First, whether the economy grows or not
in the steady state depends on the fundamentals (such as the discount factor, the total factor
productivity and the degree of substation between intermediate goods) as well as on the
18
subsidy rates. Given the fundamentals, the higher the rates of both subsidies, the more likely
the economy moves beyond the critical ck toward the balanced growth path. In fact, given
( )* 1 1k Aσβ σ= − , sufficient subsidization ensures * (1 ) (1 )c x nk k s sσ> = − − , while
** (1 ) (1 )c x nk k s sσ> = − − remains valid for all permissible rates of subsidies in the full
range of [0,1). That is, sufficient subsidization can rule out the neoclassical steady state in
the long run and replace it by the steady state with balanced growth in capital and the
variety of intermediates.
Now, we investigate the stability of the steady state by examining the asymptotic
behavior of t t tk K N= , from any initial state0 0 0 0k K N= > . The mapping 1( )t tk k −= Φ in
equation (26) is continuous: It is increasing in the range of ( )0, ck and it may be increasing
or decreasing in the range of ( ),ck ∞ .
When ( ) ( )1 1 1t c x nk k s sσ
− ≤ = − − , there is no innovation in period t, that is, 1t tN N −= .
Consequently, the two kinds of innovation-oriented subsidies are non-operative in this
neoclassical growth regime without innovation. On the other hand, when
( ) ( )1 1 1t c x nk k s sσ
− > = − − , new intermediates are introduced and both subsidies can be
operative. Economic growth in this region is driven by both the accumulation of capital and
the innovation of new varieties of intermediates. Intuitively, when the growth rate of capital
accumulation exceeds the growth rate of the variety of intermediates, the resultant ratio of
capital per variety t t tk K N= will increase; conversely, it will decrease. The slope of the
transition curve of 1( )t tk k −= Φ in this region with innovation plays a crucial role in
determining the asymptotic behavior of tk and thus deserves careful investigation.
Without the use of subsidies at 0x ns s= = , the dynamics of 1( )t tk k −= Φ in (26) will
19
become exactly the same as in Matsuyama (1999), where 1ck = , and the mapping for tk is
always decreasing in the policy-active region with innovation. Under the empirically
plausible conditions 1 � � � 1 in his benchmark model, period-2 cycles are prevalent
when tk alternates between the two regions forever. With the subsidies, it is important to
ask whether the subsidies can change the slope of the transition equation 1( )t tk k −= Φ so as
to mitigate or even eliminate the cycles.
Proposition 2. Suppose 2.θ > Define ( )0 1 1G Aβ σ≡ − and ( ) ( )11 1 cG A k σβ σ≡ − ,
where ( ) ( )1 1c x nk s sσ= − − , with 0 , 1n xs s≤ < .
(1) If 1G ≤ , then, for any given 0 0k > , the economy will eventually converge toward a
stable neoclassical stationary path with *lim t tk k→∞ = and settle down in the
policy-dormant region.
(2) If 1G >
and 1xs σ≤ , for any given 0 0k > , there may exist cycles forever if the
subsidy rates are low enough; if the subsidy rates are high enough (e.g.
( 2) / [1 ( 2)] at 0n xs sσ θ σ θ> − + − = or 1/ at 0x ns sσ→ = ), then 1| / | 1t tdk dk − <
and the economy will converge toward a stable balanced growth path oscillatorily
with lim t tk k ∗∗→∞ = .
(3) If 1G > and 1xs σ> , for any given 0 0k > , the economy will converge toward a
stable balanced growth path monotonically with lim t tk k ∗∗→∞ = .
Proof. In case (1) with ( )* 1 1 (1 ) (1 )c x nk A k s sσ σβ σ= − < = − − and 1t ck k− < , the slope
of ( ) ( )1 1
1 1( ) 1 1t t tk k A kσβ σ −
− −= Φ = − in (26) is always positive, exceeding 1 at the origin
20
( 1 0tk − → ) and falling below 1 at the steady state ( ( )*1 1 1tk k A
σβ σ− = = − ) according to:
( ) 1/1
1
(1 1/ ) 1 1tt
t
dkAk
dkσσ β σ −
−−
= − −
because 1σ > . The steady state level * ck k< is thus stable and the sequence 0{ }t tk ∞=
converges toward *k for any 0 0k > as in the standard neoclassical growth model. We
illustrate case (1) in Figure 1.
In cases (2) and (3) with ( )* 1 1 (1 ) (1 )c x nk A k s sσ σβ σ= − > = − − , the slope of the
transition equation 1( )t tk k −= Φ in (26) for 1t ck k− > is derived as
( ) 1 1/ 1
21 1
1 1 (1 ) [1 (1 1/ )][1 (1 ) ]
{1 (1 1/ ) [ (1 ) (1 )]}n n xt
t n t x n
A s s sdk
dk s k s s
σ σ
σ
β σ σ θσ θ
− −
− −
− − − − − −=
− − + − − −.
Here, 1 (1 1/ ) 0ns σ− − > because [0,1)ns ∈ and 1σ > . Also, 1 (1 ) (1 ) 0t x nk s sσ− − − − > for
1t ck k− > . So the sign of 1/t tdk dk − is the same as the sign of 11 (1 )xs σθ −− − . Recalling
1(1 1/ ) 1σθ σ −= − > under 1σ > , we have: sign 1/t tdk dk − >0 if and only if 1 1/xs σ> >
because with [0,1)xs ∈ , 11 (1 ) 0xs σθ −− − > corresponds to 1 1/xs σ> > . Accordingly,
1/ 0t tdk dk − ≤ if and only if 0 1/xs σ≤ ≤ under which 11 (1 ) 0xs σθ −− − ≤ . Also, the
absolute value of 1/t tdk dk − is monotonically decreasing in 1tk − , and approaches zero
when 1tk − approaches infinity, implying that the dynamic system in (26) cannot converge
outward to infinity.
For 1t ck k− > , there are thus two possibilities with either 1/ 0t tdk dk − ≤ or
1/ 0t tdk dk − > . If 0 1/xs σ≤ ≤ and thus 1/ 0t tdk dk − ≤ beyond ck , then the economy
may either oscillate forever with cycles or eventually converge toward the steady state of
the balanced growth path such that **lim t tk k→∞ = , depending on whether
1| / | 1 or 1t tdk dk − ≥ < . Specifically, if the subsidy rates are low enough (say zero), then
21
1| / | 1t tdk dk − > prevails under 2θ > and the economy behaves as in the original model
of Matsuyama (1999) with endogenous cycles forever. If the subsidy rates are high
enough, then we show 1| / | 1t tdk dk − < for 1/ 0t tdk dk − ≤ as follows. First, the slope
1/t tdk dk − of 1( )t tk k −= Φ at **1tk k− = can be rewritten as:
( )**
1
1 1/ 1
** 21
1 1 (1 ) [1 (1 1/ )][1 (1 ) ]|
{1 (1 1/ ) [ (1 ) (1 )]}t
n n xtk k
t n x n
A s s sdk
dk s k s s
σ σ
σ
β σ σ θσ θ−
− −
=−
− − − − − −=
− − + − − −
( )
( ){ }1 1/ 1
21 1/ 1/ 2
4 1 1 (1 ) [1 (1 1/ )][1 (1 ) ]
1 (1 1/ ) (1 ) (1 ) 1 1 (1 )
n n x
n x n n
A s s s
s s s A s
σ σ
σ σ
β σ σ θ
σ θ β σ
− −
−
− − − − − −=
− − − − − + − − + ∆,
using the expression of **k given in Proposition 1 for substitution.
For the special case without any subsidization, we have ( )0 1 1G G Aβ σ= = − and
the following
**1
1
1| 0, at 0.
t
tx nk k
t
dks s
dk G
θ− =
−
−= < = =
The absolute value of this slope exceeds one (unstable **k ) if and only if 1 1G θ< < − as
in the original model of Matsuyama (1999). This condition applies under 2θ > .
For 0 1/xs σ≤ ≤ , showing **1
1| / | 1t
t t k kdk dk
−− =
< is equivalent to showing
( ){ }( )
21 1/ 1/ 2
1 1/ 1
1 (1 1/ ) (1 ) (1 ) 1 1 (1 )
4 1 1 (1 ) [1 (1 1/ )][1 (1 ) ] 0
n x n n
n n x
F s s s A s
A s s s
σ σ
σ σ
σ θ β σ
β σ σ θ
−
− −
≡ − − − − − + − − + ∆
+ − − − − − − >
whereby 11 (1 ) 0xs σθ −− − ≤ . Using the expression for ∆ in F leads to
( )( )
( )( )
1 1/ 2
1/ 2 1 1/
1 1/ 1
2 1 1/
[1 (1 1/ ) (1 ) (1 ) 1 1 (1 ) ]
2 [1 (1 1/ ) (1 ) (1 ) 1 1 (1 ) ]
4 1 1 (1 ) [1 (1 1/ )][1 (1 ) ]
2[1 (1 1/ ) (1 ) (1 )] 2[ 1 1 (1 )
n x n n
n x n n
n n x
n x n n
F s s s A s
s s s A s
A s s s
s s s A s
σ σ
σ σ
σ σ
σ σ
σ θ β σ
σ θ β σ
β σ σ θ
σ θ β σ
−
−
− −
−
= − − − − − + − − + ∆ +
∆ − − − − − + − − +
− − − − − −
= − − − − − + − −
( )
( )
2
1 1/ 1
1 1/ 2
1 1/
]
4 1 1 (1 ) { (1 ) [ (1 ) / ] [1 (1 1/ )]
[1 (1 ) ]} 2 [1 (1 1/ ) (1 ) (1 )
1 1 (1 ) ].
n x x n n n
x n x n
n
A s s s s s s
s s s s
A s
σ σ
σ σ
σ
β σ θ σ σ
θ σ θβ σ
− −
−
−
+
− − − − + + − −
− − + ∆ − − − − − +
− −
22
Here, 1 1{ (1 ) [ (1 ) / ] [1 (1 1/ )][1 (1 ) ]}x x n n n xs s s s s sσ σθ σ σ θ− −− − + + − − − − can be shown to
be equal to [1 (1 1/ ) (1 ) (1 )]n x ns s sσσ θ− − − − − . Thus, we have
( )( )
( )( )
2 1 1/ 2
1 1/
1/ 2 1 1/
1 1/
2[1 (1 1/ ) (1 ) (1 )] 2[ 1 1 (1 ) ]
4 1 1 (1 ) [1 (1 1/ ) (1 ) (1 )]
2 [1 (1 1/ ) (1 ) (1 ) 1 1 (1 ) ]
2[1 (1 1/ ) (1 ) (1 ) 1 1 (1 )
n x n n
n n x n
n x n n
n x n n
F s s s A s
A s s s s
s s s A s
s s s A s
σ σ
σ σ
σ σ
σ σ
σ θ β σ
β σ σ θ
σ θ β σ
σ θ β σ
−
−
−
−
= − − − − − + − − +
− − − − − − − +
∆ − − − − − + − −
= − − − − − + − −
( )
2
1/ 2 1 1/
]
2 [1 (1 1/ ) (1 ) (1 ) 1 1 (1 ) ].n x n ns s s A sσ σσ θ β σ −
+
∆ − − − − − + − −
A sufficient yet unnecessary condition for 0F > is
( ) 1 1/[1 (1 1/ ) (1 ) (1 ) 1 1 (1 ) ]
[1 (1 1/ ) (1 ) (1 ) (1 )(1 )] (under 1)
0
n x n n
n x n x n
s s s A s
s s s s s G
σ σ
σ
σ θ β σ
σ θ
−− − − − − + − −
> − − − − − + − − >>
.
This condition is satisfied by the stated conditions on the subsidy rates:
( 2) / [1 ( 2)] (0,1) under >2 at 0;n xs sσ θ σ θ θ> − + − ∈ = or 1/ at 0.x ns sσ→ =
Namely, if the subsidy rates are sufficiently high such that **1
1| | 1t
t t k kdk dk
−− =
< , then the
economy will eventually converge toward the stable steady state, or the stable balanced
growth path. We depict case (2) in Figures 2 and 3.
Finally, if 1/xs σ> , showing **1
1 | 1t
t t k kdk dk
−− =
< for a stable steady state **k is
equivalent to showing the following,
( ){ }( )
21 1/ 1/ 2
1 1/ 1
1 (1 1/ ) (1 ) (1 ) 1 1 (1 )
4 1 1 (1 ) [1 (1 1/ )][1 (1 ) ] 0
n x n n
n n x
s s s A s
A s s s
σ σ
σ σ
σ θ β σ
β σ σ θ
−
− −
− − − − − + − − + ∆
− − − − − − − >
whereby 11 (1 ) 0xs σθ −− − > . The left-hand side of this inequality can be further
decomposed into
( )( )
[ ] ( )( )
2 21 1/
1 2 1 1/
1 1/ 1 1
1 1 1/
1 (1 1/ ) 1 1 (1 ) (1 ) (1 )
2 1 (1 1/ ) (1 ) (1 ) 1 1 (1 )
2 1 (1 1/ ) (1 ) (1 ) 1 1 (1 )(1 )
2 1 1 (1 ) (1 )
n n x n
n x n n
n n x x n
x n n
s A s s s
s s s A s
s s s A s s
A s s s
σ σ
σ σ
σ σ σ
σ σ
σ β σ θ
σ θ β σ
θ σ β σ
θβ σ σ
−
−
− −
− −
− − − − − + − − + ∆
+ ∆ − − − − − + − −
+ − − − − − − − −
+ − − − [ ](1 )x ns s+ −
,
which is strictly positive under 1G > , 0 1 and 1/ 1n xs sσ≤ < < < . We illustrate case (3)
23
in Figure 4. Q.E.D.
According to Proposition 2, both types of subsidies can eventually eliminate cycles
once their rates are set high enough such that 1| / | 1t tdk dk − < under which the balanced
growth path with innovation becomes stable. This is achieved either through strengthening
the demand for new intermediates (via a higher xs ) or through reducing the innovation cost
(via a higher ns ) such that R&D activities are profitable even at a low capital-variety ratio.
By increasing varieties, the subsidization can exert different impacts on final output and
thus on capital investment with a constant saving rate. First, it can directly increase final
output by increasing the number of varieties according to equation (11). Second, it can
indirectly reduce final output by reducing the equilibrium quantity of each type of
intermediate input, as the subsequent increase in the total fixed innovation cost competes for
the given amount of existing capital. For 0 1/xs σ≤ < , a higher xs increases the
equilibrium quantity of each new intermediate relative to each old intermediate, thereby
leading to a smaller indirect effect on final output as opposed to the indirect effect of a
higher ns .
It is convenient to look at how these effects work for the stability of the dynamic
system by using a general expression for the slope of the transition curve 1( )t tk k −= Φ :
21 1 1 1/ ( / ) / [ ( / ) ( / )] /t t t t t t t t t t t tdk dk d K N dk N dK dk K dN dk N− − − −= = − .
Rewrite it as the following ways that may help our interpretation:
24
1 1 1
1 1
1 1
1 1 1
1
1 1
.
t t tt
t t t t
t t t
t t t t t
t t t t t
t t t t t
dk dK dNk
dk dk dk N
dK dN K
dk K dk N N
dK k dN k k
dk K dk N k
− − −
− −
− −
− − −
= −
= −
= −
The sign of 1/t tdk dk − is determined by the terms in the bracket on the right-hand side. It
is positive (negative) if the ratio of the derivative of capital investment to the derivative
of variety expansion with respect to the initial abundance of capital,
1 1( / ) / ( / )t t t tdK dk dN dk− − , is greater (smaller) than the resultant capital-variety ratio, tk .
The absolute value of 1/t tdk dk − depends positively on the difference between the two
responses, as fractions of their new stocks, 1 1[( / ) / ( / ) / ]t t t t t tdK dk K dN dk N− −− , as well
as on the new capital-variety ratio, /t tK N . In the steady state with 1t t ck k k− = > , both
the sign and the magnitude of 1/t tdk dk − will depend solely on the gap in the respective
elasticity of tK and tN with respect to 1tk − .
Define ( )0 1 1G Aβ σ= − that serves as the growth factor in the absence of
subsidization. Holding 1tN − constant, if 1t ck k− > , from (14) and (22) we have
1 1/1 0 1/ (1 ) / [1 (1 1/ )] 0t t n t ndK dk G s N sσ σ−
− −= − − − > , while from (23) we obtain
1 1/ / [1 (1 1/ )] 0t t t ndN dk N sθ σ− −= − − > . Here, both capital investment and variety
expansion respond positively to the initial abundance of capital per variety.
Under the assumption 01 1G θ< < − , however, in the absence of subsidization the
variety response to the initial abundance of capital is more elastic than the investment
response, causing instability of the balanced growth path in the original Matsuyama
model. Without subsidization, at the steady state **k on the balanced growth path,
25
Proposition 1 and equation (23) lead to
** 00 1 0
11 for 1 and 0; /x n t t
Gk G s s N N G
θθ −− += > > = = = .
So 1 1 1/ [( / ) ( / )] /t t t t t t t tdk dk dK dk k dN dk N− − −= − at the steady state **k on the balanced
growth path equals **1
**1 0 1 0/ | ( ) / (1 ) /
tt t t tk k
dk dk G k N N Gθ θ−
− −== − = − , which is negative
under 1 θ< and smaller than 1− under 0 1G θ< − and2 θ< . In this case, period-two
cycles prevail and persist forever.
Subsidizing the fixed innovation cost strengthens the response of variety expansion to
the initial abundance of capital, 1 1/ / [1 (1 1/ )] 0t t t ndN dk N sθ σ− −= − − > , but weakens the
response of investment, 1 1/1 0 1/ (1 ) / [1 (1 1/ )] 0t t n t ndK dk G s N sσ σ−
− −= − − − > , when setting
0xs = . From (26), if this subsidy is large enough, at least for 1/ (2 1/ ) (0,1)ns σ> − ∈ , a
further increase in ns will also lead to lower capital per variety tk as long as
1 and t c t ck k k k−> > , because beyond the level 1/ (2 1/ ) (0,1)ns σ= − ∈ the numerator of
tk starts to decrease with ns . Combining them together, the sign of
1 1/1 0 1/ [ (1 ) ]( / ) / [1 (1 1/ )]t t n t t t ndk dk G s k N N sσ θ σ−
− −= − − − − is only determined by the
factor 1 1/0[ (1 ) ]n tG s kσ θ−− − whereby both terms, 1 1/
0(1 )nG s σ−− and tk θ , eventually decline
with the subsidy rate on the innovation cost when the subsidy rate becomes large enough.
This helps to explain why the sign of 1/t tdk dk − remains negative for all levels of the
subsidy rate in the regime with innovation. The remaining factors that only determine the
magnitude, not the sign, of 1/t tdk dk − are decreasing with ns as well:
11 1( / ) / [1 (1 1/ )] {1 (1 1/ ) [ (1 )]}t t n n t nN N s s k sσ σ θ −
− −− − = − − + − − for 1θ > .
This explains why the absolute value of 1/t tdk dk − becomes smaller when the subsidy rate
ns becomes larger.
26
Subsidizing the purchase of new intermediates does not affect the responses of capital
investment and variety expansion to the initial abundance of capital, when setting 0.ns =
The sign of 1/t tdk dk − is merely determined by 0[ ]tG k θ− which is initially negative when
the subsidy rate is equal to zero. It follows from (26) that a higher subsidy rate for the
purchase of new intermediates will reduce the amount of capital per variety tk as long as
1 and t c t ck k k k−> > , because /t xdk ds is signed by 1[1 (1 ) ( 1)] 0x x ts s kσσ θ θ σ−− − − − − <
in which 1 (1 ) 0x xs s σσ θ− − − < attains a maximum 1 / 1 0σ − < at 1/ .xs σ=
Consequently, a higher subsidy rate for the purchase of new intermediates will increase the
value of 0[ ]tG k θ− in general or reduce the absolute value 0| |tG k θ− when 0 0tG k θ− < in
particular. At **0G k θ= as a special case on the balanced growth path, the corresponding
subsidy rate is 1/xs σ= that leads to **1
1/ | 0t
t t k kdk dk
−− =
= . When the subsidy rate is
increased further for 1/xs σ> , **0[ ] 0G k θ− > must hold, leading to 1/ 0t tdk dk − > on the
balanced growth path. Recall that subsidizing the purchase of new intermediates at a rate
1/xs σ> will lead to c mx x< , thereby creating a loss in final output. The loss in final
output, due to a higher xs beyond 1/xs σ= , will in turn lead to a decline in capital
investment for a constant saving rate given any initial 1tk − , while variety expansion
accelerates at a higher xs . Consequently, a higher xs with 1/xs σ> will reduce **k to a
level such that **0[ ] 0G k θ− > and thus 1/ 0t tdk dk − > at the steady state. For 0ns = , the
absolute value of 1/t tdk dk − on the balanced growth path is derived below:
**1
1/20
1/21 0
1 (1 )| 1
1 (1 )t
t xk k
t x
dk G s
dk G s
σ
σθθ− =
−
+ − − − ∆= <+ − − + ∆
.
Therefore, the balanced growth path becomes stable once the subsidy rate on the
purchase of new intermediates is set high enough such that **1
1| / | 1t
t t k kdk dk
−− =
< .
27
The analysis so far has little quantitative implications of the subsidies and sheds no
light on the welfare consequences of the subsidies and on the optimal subsidy rates due to
the model's complexity. To gain insights in these directions, we now turn to numerical
simulations.
4. Numerical simulation results
In this section, we will gauge the quantitative implications and the welfare gains from the
subsidies and find optimal subsidy rates by numerical simulations for plausible
parameterizations.
Since the two kinds of subsidies can promote growth, investment and innovation and
since they can eliminate cyclical fluctuations, they have potential for enhancing social
welfare compared with the suboptimal benchmark model without any government
subsidization. Starting with lower final output and slower innovation and growth under
monopoly pricing compared to their socially optimal levels in the Romer regime (as shown
in Barro and Sala-i-Martin, 1995), increasing the subsidies may have different impacts on
final output on the one hand and on innovation and growth on the other. Thus, increasing
the subsidy rates tends to have opposing impacts on welfare. The positive effect of subsidies
on innovation and growth tends to enhance welfare when the innovation rate and the growth
rate are lower than their socially optimal levels. The effect of subsidies on final output is
initially positive at low subsidy rates and eventually negative at sufficiently high subsidy
rates, as shown earlier in our model.
Moreover, when subsidies mitigate cyclical fluctuations in consumption, investment
28
and innovation, there are possible efficiency gains due to diminishing marginal utility and
diminishing marginal product. Thus, the overall welfare effect is expected to be initially
positive, when the subsidy rates are low, but eventually negative, when the subsidy rates
become high enough. Unlike existing studies of R&D subsidization that focus on the
balanced growth path only, the model here allows us to explore whether the optimal rates of
subsidies lie in the range where they eventually eliminate cycles with oscillatory or
monotonic convergence toward the steady state on the balanced growth path.
We set a benchmark parameterization as=2.5, =0.63, =2.4414, and =5A β θ σ . The
value of 5σ = is in line with that in Matsuyama (1999) where it plays dual roles:
1 1 0.8σ− = is the share of capital (interpreted broadly as both physical and human capital);
( )1/ 1 0.25σ − = is the monopoly mark-up enjoyed by the innovator. When one period (the
length of patent protection) corresponds to 15 years3, an annual discount factor of 0.97 leads
to ( )150.97 0.63β = = .
The value of A is chosen such that the calibrated economy starts with period-2 cycles
in the absence of subsidization. According to Proposition 2, the benchmark parameterization
without subsidies, 0xs = and 0ns = , satisfies1 1G θ< < − , indicating that the economy
grows through period-2 cycles, perpetually moving back and forth between the two regimes.
This is an empirically plausible case as argued in Matsuyama (1999). Moreover, we choose
the initial states as 0 0.4K = and 0 1N = .
We track down the equilibrium dynamics and calculate the infinitely lived
3 According to WTO’s Agreement on Trade-Related Aspects of Intellectual Property Rights, the term of patent is 20 years from the filing date of the application. However, many countries have laws with shorter terms of 6 to10 years. In our simulation, we take the average term of patent as 15 years. In fact, the shorter the term of patents, the larger the discount factor, thereby the stronger the positive results from subsidization.
29
representative agent’s welfare in (15) by using the first-order difference equations in (22)
and (23) to update the states and t tK N and then finding final output, consumption, the tax
rate and utility in every period. In order to better comprehend the change in welfare, we use
its equivalent variation in consumption in every period. Define the equivalent variation in
consumption in each period by ∆ that allows the benchmark case without subsidization to
reach the same welfare level as that in a case with subsidization type i:
no-subsidization ln(1 )1 isU U
ββ
+ + ∆ =−
.
This corresponds to adding ( ) ( ) ( )1
ln 1 ln 1 1t
tβ β β∞
=+ ∆ = + ∆ − ∑ to the welfare level in
the benchmark case without subsidization. We work out a large number of periods in each
case (say 1000) such that any further increase in the number of periods has no impact on
welfare within ten decimal digits.
In Table 1 and Table 2, we report the simulation results when increasing xs and ns
from zero to reach and go beyond a peak of the welfare level, one at a time, respectively. In
Figure 5 and Figure 6, we also depict the capital variety ratios in the two regimes( ),H Lk k in
the scenario of period-2 cycles or the steady state capital variety ratio, k∗∗ , on the balanced
growth path when increasing xs and ns from zero to sufficiently high rates. The
subsequent welfare levels are plotted in the same figure to better illustrate the range in
which the optimal rates of subsidies can maximize welfare.
In Table 1, we set 0ns = and examine the dynamic behavior, the balanced growth rate,
the consumption tax rate, the welfare level, and the consumption equivalent variation when
varying xs from 0 in the benchmark case to 40% gradually (by 0.001 each step). When xs is
sufficiently small (e.g., 0.01xs = ) the economy still alternates between the policy-dormant
30
and the policy-active regions, and the social welfare is higher than that in the case without
subsidization. When xs is increased further but still below 1/ 0.2σ = , the steady state k ∗∗
becomes stable and the economy achieves oscillatory convergence toward k ∗∗ as predicted
in case 2 of Proposition 2. As xs is increased beyond 1 0.2σ = , the condition for case 3 of
Proposition 2 is satisfied for tk to converge toward k ∗∗ monotonically. For a better view of
the welfare effect, in Figure 5 we vary xs from 0 to 0.6 ( 0ns = ) and find the subsequent
welfare levels. It is worth noting that the welfare level is a concave and smooth curve,
peaking at a unique point when 0.16xs = , due to a balance between the various gains and
losses in efficiency mentioned earlier. This optimal rate lies in the range where the subsidy
eliminates the cycles with oscillatory convergence toward the steady state. The resultant
welfare level at the optimal subsidy rate is 0.0643 compared to -0.0811 in the benchmark
case without subsidization. The change in welfare is equivalent to a significant 8.91% rise
in consumption in every period from the benchmark case without subsidization.
In Table 2 and Figure 6, we fix 0xs = and focus on the effects of the subsidy on the
fixed R&D cost, ns . Period-2 cycles persist for a relatively wide range of ns . After that,
when it is high enough to satisfy the condition for case 2 of Proposition 2, period-2 cycles
are replaced by oscillatory convergence to k ∗∗ in the policy-active region. In Figure 6, we
vary ns from 0 to 90% ( 0xs = ) and find the subsequent welfare level. It is worth noting
that the welfare curve is concave and smooth as well, peaking at a unique point when
* 0.61ns = that achieves oscillatory convergence toward the steady state with balanced
growth in capital and variety. The welfare curve is flatter before peaking and takes longer to
reach the optimal level of the subsidy rate than in Figure 5, because this subsidy ns does
31
not change the price gap and therefore does not create the additional gain or loss in static
efficiency as those created by xs on either side of 1/xs σ= . So the efficiency gain from
faster variety expansion and from convergence to the steady state at a higher subsidy rate to
the fixed R&D cost is gradually offset by a loss from the subsequent decline in the
equilibrium quantity of each intermediate. Beyond this optimal subsidy rate, the welfare
level declines rapidly, because the efficiency loss from the declined use of each intermediate
is increasing at the margin. The maximized welfare level is 0.0668 against -0.0811 in the
benchmark case without subsidization. The welfare change is equivalent to a significant
9.07% rise in consumption in every period.
In Figure 7, we jointly use the subsidy to the new intermediate goods and the subsidy
to R&D investment at the same time and plot the welfare surface when varying both xs and
ns from zero to sufficiently high rates. It turns out the welfare surface is concave and smooth,
peaking at a unique point when 0.1, and 0.38x ns s= = . The capital variety ratio will
converge to the steady state, 0.4972k∗∗ = , oscillatorily in the policy-active regime when
the subsidies are financed by a flat-rate consumption tax of 9.21%. The maximized welfare
level is 0.0866, higher than those in cases with one subsidy at a time as expected. The
welfare change from the benchmark case is equivalent to a substantial 10.35% rise in
consumption in each period.
Note that the respective optimal rates of subsidies mitigate and eventually eliminate the
cycles in all the reported cases. So part of the welfare gain must have come from smoothing
consumption, which is new compared to the results in existing analysis of subsidization on
the balanced growth path alone.
32
5. Conclusion
We have examined the implications of two types of subsidies, one type to the purchase of
new intermediate products and the other to R&D investment, in the model of Matsuyama
(1999) with growth through endogenous cycles. One contribution of doing so is that the
subsidization can reduce the critical level of the capital-variety ratio substantially,
enhancing the possibility for the economy to stay at the policy-active region with
sustainable innovation and growth. Sufficient subsidization can rule out the neoclassical
regime without innovation from the steady state in the long run.
Another contribution is that we have characterized several possible scenarios of the
asymptotic paths of the representative agent economy from any initial state depending on
the values of the economic fundamentals and subsidy rates. In a novel yet empirically
plausible scenario, sufficient subsidization stabilizes the balanced growth path with
innovation and thus leads to oscillatory or monotonic convergence towards it.
Also, we have used numerical examples to gauge the welfare gains from the two types
of subsidies starting from an empirically plausible parameterization for a period-2 cycle
economy in the absence of subsidization. It turns out that both types of subsidies can help
enhance the welfare level significantly in terms of a 9-10% rise in consumption for each
period; and the optimal subsidy rates maximizing social welfare are calculated in their
plausible ranges that eventually eliminate cycles for consumption smoothing.
Our results in this paper appear consistent not only with substantial subsidization to
new investment and R&D spending but also with the combination of intensified innovation
33
and dampened cyclical fluctuations in many industrial nations in the postwar era.
Our results about R&D subsidies are different from those in Haruyama (2009): optimal
R&D subsidies cannot eliminate cycles in their model but can do so in our model. As
mentioned earlier, the models are very different (intermediates production and innovation
must use accumulated capital in the present model but only use labor in their model). So our
results concerning the R&D subsidies are complementary to their results. In particular, our
results about the subsidies to the purchase of new intermediates are new in models with
endogenous cycles and sustainable growth.
34
References
Aghion, P., Howitt, P., 1992. A model of growth through creative destruction.
Econometrica 60, 323-351.
Aloi, M., Lasselle, L., 2007. Growth and welfare effects of stabilizing innovation cycles.
Journal of Macroeconomics 29, 806-823.
Barro, R.J., Sala-i-Martin, X., 1995. Economic Growth, McGraw-Hill, New York.
Bental, B., Peled, D., 1996. The accumulation of wealth and the cyclical generation of new
technologies: A search theoretic approach. International Economic Review 37 (3), 687-718.
Cass, D., 1965. Optimum growth in an aggregative model of capital accumulation. Review
of Economic Studies 32, 233-240.
Fan, J., 1995. Endogenous technical progress, R&D periods and durations of business
cycles. Journal of Evolutionary Economics 5, 341-368.
Francois, P., Lloyd-Ellis, H., 2003. Animal spirit through creative destruction. American
Economic Review 93 (3), 530-550.
Francois, P., Shi, S., 1999. Innovation, growth, and welfare-improving cycles. Journal of
Economic Theory 85, 226-257.
Freeman, S., Hong, D., and Peled, D., 1999. Endogenous cycles and growth with indivisible
technological developments. Review of Economic Dynamics 2, 403-432.
Gordon, R., Kalambokidis, L., Rohaly, J., Slemrod, J., 2004. Toward a consumption tax and
beyond. American Economic Review 94, 161-165.
Gordon, R., Kalambokidis, L., Slemrod, J., 2004. Do we now collect any revenue from
taxing capital income? Journal of Public Economics 88, 981-1009.
35
Grossman, J., Helpman, E., 1991. Quality ladders in the theory of growth. Review of
Economic Studies 58, 43-61.
Haruyama, T., 2009. R&D policy in a volatile economy. Journal of Economic Dynamics and
Control 33, 1761-1778.
Koopmans, T. C., 1965. On the concept of optimal economic growth. The Econometric
Approach to Development Planning, Amsterdam, North Holland, 1965.
Kydland, F.E., Prescott, E.C., 1982. Time to build and aggregate fluctuations.
Econometrica 50, 1345-1370.
Long, J.B., Plosser, C.I., 1983. Real business cycles. Journal Political Economy 91, 39-69.
Maliar, L., Maliar, S., 2004. Endogenous growth and endogenous business cycles.
Macroeconomic Dynamics 8, 559-581.
Matsuyama, K., 1996. Growing through cycles, CMS-EMS Working Paper #1203,
Northwestern University.
Matsuyama, K., 1999. Growing through cycles. Econometrica 67, 335-347.
Matsuyama, K., 2001. Growing through cycles in an infinitely lived agent economy.
Journal of Economic Theory 100, 220-234.
Parente, S.L., 1994. Technology adoption, learning-by-doing, and economic growth. Journal
of Economic Growth 63, 346-369.
Romer, P., 1990. Endogenous technological change. Journal of Political Economy 98,
S71-S102.
Segerstrom, P., 1991. Innovation, imitation, and economic growth. Journal of Political
Economy 99, 807-827.
36
Solow, R., 1956. A contribution to the theory of economic growth. Quarterly Journal of
Economics 70, 65-94.
Swan, T., 1956. Economic growth and capital accumulation. Economic Record 32, 334-361.
Wlade, K., 2002. The economic determinants of technology shocks in a real business cycle.
Journal of Economic Dynamics and Control 27, 1-28.
Walde, K., 2005. Endogenous growth cycles. International Economic Review 46 (3),
867-894.
Young, A., 1993. Invention and bounded learning by doing. Journal of Political Economy
101, 443-472.
Zeng, J., Zhang, J., 2007. Subsidies in an R&D growth model with elastic labor,
Journal of Economic Dynamics and Control 31, 861-886.
37
Table 1. Results of changing the subsidy to the purchase of new intermediate goods
Benchmark parameterization: 0 0A=2.5, =0.63, =2.4414, =5, 0.4, 1, 0n xK N s sβ θ σ = = = =
Subsidy Mode of k ∗∗ or Growth rate Tax rate Welfare C equivalent
rates dynamics ( ),L Hk k gross (annual %) % level variation %
0xs =
Period-2 cycle (0.983, 1.243) 1.262 (1.56) 0.0, 0.0 -0.0811 0.0
0.01xs = Period-2 cycle (0.945, 1.204) 1.272 (1.62) 0.0, 0.34 -0.0716 0.56
0.02xs = Oscillatory 1.020 1.283 (1.67) 0.37 -0.0626 1.09
0.05xs = Oscillatory 0.904 1.317 (1.85) 1.17 -0.0292 3.10
0.10xs = Oscillatory 0.743 1.371 (2.13) 3.42 0.0226 6.28
0.15xs = Oscillatory 0.616 1.420 (2.37) 7.25 0.0631 8.84
* 0.16xs = Oscillatory 0.594 1.429 (2.41) 8.27 0.0643 8.91
0.20xs = Monotonic 0.516 1.460 (2.56) 13.35 0.0482 7.89
0.25xs = Monotonic 0.437 1.488 (2.68) 22.60 -0.0624 1.10
0.30xs = Monotonic 0.374 1.503 (2.75) 36.32 -0.2762 -12.14
0.35xs = Monotonic 0.323 1.504 (2.76) 56.60 -0.5908 -34.90
0.40xs = Monotonic 0.280 1.493 (2.71) 87.23 -1.0077 -72.32
Note: (1) The growth rate in the period-2 cycle economy is calculated as the geometric average of the corresponding growth rates in the two regions following Matsuyama (1999). (2) The values in the brackets beside growth rates indicate the discounted annual rates. (3) The subsidy rates with∗ indicates the optimal rate maximizing the social welfare.
38
Table 2. Results of changing the subsidy to the R&D investment
Benchmark parameterization: 0 02.5, 0.63, 2.4414, 5, 0.4, 1, 0x nA K N s sβ θ σ= = = = = = = =
Subsidy Mode of k ∗∗ or Growth rate Tax rate Welfare C equivalent
rates dynamics ( ),L Hk k gross (annual %) % level variation %
0ns = Period-2 cycle (0.983, 1.243) 1.262 (1.56) 0.0, 0.0 -0.0811 0.0
0.1ns = Period-2 cycle (0.888, 1.146) 1.286 (1.69) 0,0, 0.76 -0.0616 1.15
0.2ns = Period-2 cycle (0.794, 1.047) 1.311 (1.82) 0.0, 1.85 -0.0370 2.62
0.3ns = Period-2 cycle (0.700, 0.947) 1.339 (1.97) 0.0, 3.44 -0.0063 4.49
0.31ns = Oscillatory 0.795 1.342 (1.98) 1.79 -0.0034 4.67
0.4ns = Oscillatory 0.703 1.371 (2.13) 2.87 0.0210 6.18
0.5ns = Oscillatory 0.600 1.407 (2.30) 4.69 0.0578 8.50
0.6ns = Oscillatory 0.495 1.446 (2.49) 7.70 0.0665 9.06
0.61ns∗ = Oscillatory 0.485 1.450 (2.51) 8.10 0.0668 9.07
0.7ns = Oscillatory 0.388 1.486 (2.68) 13.15 0.0141 5.75
0.8ns = Oscillatory 0.277 1.523 (2.85) 24.94 -0.2787 -12.31
0.9ns = Oscillatory 0.159 1.518 (2.82) 63.02 -1.5241 -133.37
Note: (1) The growth rate in the period-2 cycle economy is calculated as the geometric average of the corresponding growth rates in the two regions following Matsuyama (1999). (2) The values in the brackets beside growth rates indicate the discounted annual rates.
(3) The subsidy rates with * indicates the optimal rate maximizing the social welfare.
39
Figure 1. 1G <
Figure 2. 1G > , 1xs σ< and **1
1| / | 1t
t t k kdk dk
−− =
>
0 0.5 1 1.50
0.5
1
1.5
k(t-1)
k(t)
k*k0 kc
0 0.5 1 1.50
0.5
1
1.5
k(t-1)
k(t)
kc k**
40
Figure 3. 1G > , 1xs σ< and **1
1| / | 1t
t t k kdk dk
−− =
<
Figure 4. 1G > and 1xs σ>
0 0.5 1 1.50
0.5
1
1.5
k(t-1)
k(t)
k0 kc k**
0 0.5 1 1.50
0.5
1
1.5
k(t-1)
k(t)
k0 kc k**
41
Figure 5. The dynamics of tk and welfare levels by changing xs
Figure 6. The dynamics of tk and welfare levels by changing ns
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7-5
-4
-3
-2
-1
0
1
2
subsidy to intermediates, sx
kH
kL
welfare
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-2.5
-2
-1.5
-1
-0.5
0
0.5
1
1.5
subsidy to R&D investment, sn
kH
kL
welfare
42
Figure 7. The welfare by using both xs and ns
00.2
0.40.6
0.8
00.2
0.40.6
0.8-7
-6
-5
-4
-3
-2
-1
0
1
snsx
wel
fare
ECONOMICS DISCUSSION PAPERS
2009 DP NUMBER
AUTHORS TITLE
09.01 Le, A.T. ENTRY INTO UNIVERSITY: ARE THE CHILDREN OF IMMIGRANTS DISADVANTAGED?
09.02 Wu, Y. CHINA’S CAPITAL STOCK SERIES BY REGION AND SECTOR
09.03 Chen, M.H. UNDERSTANDING WORLD COMMODITY PRICES RETURNS, VOLATILITY AND DIVERSIFACATION
09.04 Velagic, R. UWA DISCUSSION PAPERS IN ECONOMICS: THE FIRST 650
09.05 McLure, M. ROYALTIES FOR REGIONS: ACCOUNTABILITY AND SUSTAINABILITY
09.06 Chen, A. and Groenewold, N. REDUCING REGIONAL DISPARITIES IN CHINA: AN EVALUATION OF ALTERNATIVE POLICIES
09.07 Groenewold, N. and Hagger, A. THE REGIONAL ECONOMIC EFFECTS OF IMMIGRATION: SIMULATION RESULTS FROM A SMALL CGE MODEL.
09.08 Clements, K. and Chen, D. AFFLUENCE AND FOOD: SIMPLE WAY TO INFER INCOMES
09.09 Clements, K. and Maesepp, M. A SELF-REFLECTIVE INVERSE DEMAND SYSTEM
09.10 Jones, C. MEASURING WESTERN AUSTRALIAN HOUSE PRICES: METHODS AND IMPLICATIONS
09.11 Siddique, M.A.B. WESTERN AUSTRALIA-JAPAN MINING CO-OPERATION: AN HISTORICAL OVERVIEW
09.12 Webe r, E.J. PRE-INDUSTRIAL BIMETALLISM: THE INDEX COIN HYPTHESIS
09.13 McLure, M. PARETO AND PIGOU ON OPHELIMITY, UTILITY AND WELFARE: IMPLICATIONS FOR PUBLIC FINANCE
09.14 Weber, E.J. WILFRED EDWARD GRAHAM SALTER: THE MERITS OF A CLASSICAL ECONOMIC EDUCATION
09.15 Tyers, R. and Huang, L. COMBATING CHINA’S EXPORT CONTRACTION: FISCAL EXPANSION OR ACCELERATED INDUSTRIAL REFORM
09.16 Zweifel, P., Plaff, D. and Kühn, J.
IS REGULATING THE SOLVENCY OF BANKS COUNTER-PRODUCTIVE?
09.17 Clements, K. THE PHD CONFERENCE REACHES ADULTHOOD
09.18 McLure, M. THIRTY YEARS OF ECONOMICS: UWA AND THE WA BRANCH OF THE ECONOMIC SOCIETY FROM 1963 TO 1992
09.19 Harris, R.G. and Robertson, P. TRADE, WAGES AND SKILL ACCUMULATION IN THE EMERGING GIANTS
09.20 Peng, J., Cui, J., Qin, F. and Groenewold, N.
STOCK PRICES AND THE MACRO ECONOMY IN CHINA
09.21 Chen, A. and Groenewold, N. REGIONAL EQUALITY AND NATIONAL DEVELOPMENT IN CHINA: IS THERE A TRADE-OFF?
ECONOMICS DISCUSSION PAPERS
2010
DP NUMBER
AUTHORS TITLE
10.01 Hendry, D.F. RESEARCH AND THE ACADEMIC: A TALE OF TWO CULTURES
10.02 McLure, M., Turkington, D. and Weber, E.J. A CONVERSATION WITH ARNOLD ZELLNER
10.03 Butler, D.J., Burbank, V.K. and Chisholm, J.S.
THE FRAMES BEHIND THE GAMES: PLAYER’S PERCEPTIONS OF PRISONER’S DILEMMA, CHICKEN, DICTATOR, AND ULTIMATUM GAMES
10.04 Harris, R.G., Robertson, P.E. and Xu, J.Y. THE INTERNATIONAL EFFECTS OF CHINA’S GROWTH, TRADE AND EDUCATION BOOMS
10.05 Clements, K.W., Mongey, S. and Si, J. THE DYNAMICS OF NEW RESOURCE PROJECTS A PROGRESS REPORT
10.06 Costello, G., Fraser, P. and Groenewold, N. HOUSE PRICES, NON-FUNDAMENTAL COMPONENTS AND INTERSTATE SPILLOVERS: THE AUSTRALIAN EXPERIENCE
10.07 Clements, K. REPORT OF THE 2009 PHD CONFERENCE IN ECONOMICS AND BUSINESS
10.08 Robertson, P.E. INVESTMENT LED GROWTH IN INDIA: HINDU FACT OR MYTHOLOGY?
10.09 Fu, D., Wu, Y. and Tang, Y. THE EFFECTS OF OWNERSHIP STRUCTURE AND INDUSTRY CHARACTERISTICS ON EXPORT PERFORMANCE
10.10 Wu, Y. INNOVATION AND ECONOMIC GROWTH IN CHINA
10.11 Stephens, B.J. THE DETERMINANTS OF LABOUR FORCE STATUS AMONG INDIGENOUS AUSTRALIANS
10.12 Davies, M. FINANCING THE BURRA BURRA MINES, SOUTH AUSTRALIA: LIQUIDITY PROBLEMS AND RESOLUTIONS
10.13 Tyers, R. and Zhang, Y. APPRECIATING THE RENMINBI
10.14 Clements, K.W., Lan, Y. and Seah, S.P. THE BIG MAC INDEX TWO DECADES ON AN EVALUATION OF BURGERNOMICS
10.15 Robertson, P.E. and Xu, J.Y. IN CHINA’S WAKE: HAS ASIA GAINED FROM CHINA’S GROWTH?
10.16 Clements, K.W. and Izan, H.Y. THE PAY PARITY MATRIX: A TOOL FOR ANALYSING THE STRUCTURE OF PAY
10.17 Gao, G. WORLD FOOD DEMAND
10.18 Wu, Y. INDIGENOUS INNOVATION IN CHINA: IMPLICATIONS FOR SUSTAINABLE GROWTH
10.19 Robertson, P.E. DECIPHERING THE HINDU GROWTH EPIC
10.20 Stevens, G. RESERVE BANK OF AUSTRALIA-THE ROLE OF FINANCE
10.21 Widmer, P.K., Zweifel, P. and Farsi, M. ACCOUNTING FOR HETEROGENEITY IN THE MEASUREMENT OF HOSPITAL PERFORMANCE
10.22 McLure, M. ASSESSMENTS OF A. C. PIGOU’S FELLOWSHIP THESES
10.23 Poon, A.R. THE ECONOMICS OF NONLINEAR PRICING: EVIDENCE FROM AIRFARES AND GROCERY PRICES
10.24 Halperin, D. FORECASTING METALS RETURNS: A BAYESIAN DECISION THEORETIC APPROACH
10.25 Clements, K.W. and Si. J. THE INVESTMENT PROJECT PIPELINE: COST ESCALATION, LEAD-TIME, SUCCESS, FAILURE AND SPEED
10.26 Chen, A., Groenewold, N. and Hagger, A.J. THE REGIONAL ECONOMIC EFFECTS OF A REDUCTION IN CARBON EMISSIONS
10.27 Siddique, A., Selvanathan, E.A. and Selvanathan, S.
REMITTANCES AND ECONOMIC GROWTH: EMPIRICAL EVIDENCE FROM BANGLADESH, INDIA AND SRI LANKA
ECONOMICS DISCUSSION PAPERS
2011 DP NUMBER AUTHORS TITLE
11.01 Robertson, P.E. DEEP IMPACT: CHINA AND THE WORLD ECONOMY
11.02 Kang, C. and Lee, S.H. BEING KNOWLEDGEABLE OR SOCIABLE? DIFFERENCES IN RELATIVE IMPORTANCE OF COGNITIVE AND NON-COGNITIVE SKILLS
11.03 Tu rkington, D. DIFFERENT CONCEPTS OF MATRIX CALCULUS
11.04 Golley, J. and Tyers, R. CONTRASTING GIANTS: DEMOGRAPHIC CHANGE AND ECONOMIC PERFORMANCE IN CHINA AND INDIA
11.05 Collins, J., Baer, B. and Weber, E.J. ECONOMIC GROWTH AND EVOLUTION: PARENTAL PREFERENCE FOR QUALITY AND QUANTITY OF OFFSPRING
11.06 Turkington, D. ON THE DIFFERENTIATION OF THE LOG LIKELIHOOD FUNCTION USING MATRIX CALCULUS
11.07 Groenewold, N. and Paterson, J.E.H. STOCK PRICES AND EXCHANGE RATES IN AUSTRALIA: ARE COMMODITY PRICES THE MISSING LINK?
11.08 Chen, A. and Groenewold, N. REDUCING REGIONAL DISPARITIES IN CHINA: IS INVESTMENT ALLOCATION POLICY EFFECTIVE?
11.09 Williams, A., Birch, E. and Hancock, P. THE IMPACT OF ON-LINE LECTURE RECORDINGS ON STUDENT PERFORMANCE
11.10 Pawley, J. and Weber, E.J. INVESTMENT AND TECHNICAL PROG RESS IN THE G7 COUNTRIES AND AUSTRALIA
11.11 Tyers, R. AN ELEM ENTAL MA CROECONOMIC MODEL F OR APPLIED ANALYSIS AT UNDERGRADUATE LEVEL
11.12 Clements, K.W. and Gao, G. QUALITY, QUANTITY, SPENDING AND PRICES
11.13 Tyers, R. and Zhang, Y. JAPAN’S ECONOMIC RECOVERY: INSIGHTS FROM MULTI-REGION DYNAMICS
11.14 McLure, M. A. C. PIGOU’S REJECTION OF PARETO’S LAW
11.15 Kristoffersen, I. THE SUBJECTIVE WELLBEING SCALE: HOW REASONABLE IS THE CARDINALITY ASSUMPTION?
11.16 Clements, K.W., Izan, H.Y. and Lan, Y. VOLATILITY AND STOCK PRICE INDEXES
11.17 Parkinson, M. SHANN MEMORIAL LECTURE 2011: SUSTAINABLE WELLBEING – AN ECONOMIC FUTURE FOR AUSTRALIA
11.18 Chen, A. and Groenewold, N. THE NATIONAL AND REGIONAL EFFECTS OF FISCAL DECENTRALISATION IN CHINA
11.19 Tyers, R. and Corbett, J. JAPAN’S ECONOMIC SLOWDOWN AND ITS GLOBAL IMPLICATIONS: A REVIEW OF THE ECONOMIC MODELLING
11.20 Wu, Y. GAS MARKET INTEGRATION: GLOBAL TRENDS AND IMPLICATIONS FOR THE EAS REGION
11.21 Fu, D., Wu, Y. and Tang, Y. DOES INNOVATION MATTER FOR CHINESE HIGH-TECH EXPORTS? A FIRM-LEVEL ANALYSIS
11.22 Fu, D. and Wu, Y. EXPORT WAGE PREMIUM IN CHINA’S MANUFACTURING SECTOR: A FIRM LEVEL ANALYSIS
11.23 Li, B. and Zhang, J. SUBSIDIES IN AN ECONOMY WITH ENDOGENOUS CYCLES OVER NEOCLASSICAL INVESTMENT AND NEO-SCHUMPETERIAN INNOVATION REGIMES
11.24 Krey, B., Widmer, P.K. and Zweifel, P. EFFICIENT PROVISION OF ELECTRICITY FOR THE UNITED STATES AND SWITZERLAND
11.25 Wu, Y. ENERGY INTENS ITY AND I TS DETERMINANTS I N CHINA’S REGIONAL ECONOMIES