Innovation policy in the Global Economy: a European Perspective

Giammario Impullitti (University of Nottingham) and Teodora Borota (Uppsala University )

SITE, February 4, 2014

1/34

Innovation and technological progress are main drivers of long-run growth and cross-country convergence

Market failures may lead to underinvestment in innovation and fiscal policy may be used to intervene

Monetary and Regulation policy are common to all EU countries, Fiscal policy is not

Is there a scope for cooperative or unified tax treatment for innovation activities in the EU?

What is the role played by countries technological/innovation asymmetries?

Can a EU-level fiscal treatment of innovation speed up convergence?

2/34

Assymetric R&D expenditure

The new members’ expenditure relative to the old members’ expenditure increased from 11.7% in 2005 to 20.7% in 2011

3/34

Rising R&D actvity in the new member states

4/34

Innovation activity in the new member states

5/34

Innovation activity in the new member states

6/34

Manufacturing industry sectors - varying innovation activity

7/34

...in percentage

8/34

FDI positions of the EU15 in the new member states increase

9/34

Correlation between FDI and the innovation activity

10/34

Innovation activity vs. foreign ownership

11/34

Non-negligible R&D performance by the new member states

Asymmetric innovation activity across the EU

Positive correlation between the FDI and innovation across sectors

12/34

Asymmetric government support to R&D

Diverse support instruments and intensities across EU states

13/34

New members have incentives to subsidize R&D to attract FDI→more jobs + spillovers→leapfrogging

Old members have incentives to subsidize R&D to keep profits and production at home

Is there incentive for policy cooperation?

14/34

Strategic trade and industrial policy (Eaton and Grossman, 1986, Neary and Leahy, 2000, Haaland and Kind, 2007) - static and/or partial equilibrium; without endogenous growth, static models of innovation and strategic R&D cannot account for the long-run effects on consumer surplus

Optimal R&D subsidies in closed economy (Jones, 2000, Segerstrom, 1999, Acemoglu, Akcigit, Bloom, Kerr, 2013) - we bring the trade-offs of the open economy set-up where international R&D competition has a strategic aspect

FDI, product cycle and innovation (Dinopoulos and Segerstrom, 2010, Acemoglu, Gancia, Zillibotti 2012) - abstract from the innovation in the follower country and strategic policy games

Tax competition (e.g. Mendoza Tesar, JME 2005)

15/34

Two-region model, West-East, innovation-driven Schumpeterian growth model

West is technological leader:

firms invest in innovative R&D in all existing sectors to get global leadership leaders can transfer production into the East (FDI) which requires some adaptive R&D West R&D subsidies for R&D in the West

East is technological follower:

Firms innovate only in sectors with West FDI→ technological leapfrogging (Theoretical Novelty)

East R&D subsidies for Eastern innovative R&D and Western firms’ adaptive R&D

Governments of West and East decide on the level of subsidies to the R&D sectors

16/34

1 Vary a region’s subsidy (other region’s subsidy fixed)→effect on innovation rates, sectors’ composition, relative wage, expenditure and welfare

2 Best response subsidies of each region when they play a noncooperative game

3 Assume some level of cooperation (or caring for other region’s welfare besides maximizing only one’s own) - is cooperation welfare improving?

4 Repeat the previous exercise for various degrees - what is the optimal degree of cooperation for EU? Is it possible to implement?

17/34

Industry composition

18/34

Households have preferences with unit elasticity over an infinite set of goods ω∈ [0, 1].

Each household supplies one unit of labor and solves max U=

Z_∞ 0

N0e^−(ρ−n)tlog u(t)dt with instantaneous utility

log u(t) ≡ Z1

0

log

"_jmax(ω,t) j

∑

λ^j(ω,t)q(j , ω, t)

# d ω

q(_{j , ω, t})is the per-member flow of good ω, of quality index j∈ {0, 1, 2, ...} A new vintage of a good ω yields a quality equal toλtimes the quality of the previous vintage (λ>1) with vintages as perfect substitutes

Consumers pay a lump-sum tax to finance the investment in R&D

19/34

2 activities - production ofgoodsandideas

Firms produce any consumption good ω∈ [0, 1]using a linear (mobile) technology

Unit labor requirement is a^k, where k=W , E , M stands for Western (W , old member), Eastern (E , new member) and multinational/West subsidiary firms (M)

The wage rate is w^K, with w^W >w^E

Each industry ω is monopolistically competitive, with the top quality product protected by patent, which becomes obsolete when a new quality vintage arrives

20/34

Leaders are challenged by the R&D firms that employ workers and produce a probability intensity of inventing the next vintage

Before the establishment of subsidiary firms, only West is capable of R&D Once a subsidiary adaptive R&D has been conducted, technology transfer makes possible for R&D in the East

The arrival rate of innovation in industry ω is the sum of the Poisson arrival rates of all R&D firms targeting product ω

I_i^K(ω, t) =^A

kι^K_i (ω, t) X(ω, t) ^,

where X(ω, t) =2κN(t)measures the degree of complexity in the innovation process

21/34

Static optimization: we assume Bertrand competition, so in equilibrium only the top quality product is produced and consumed

q(j , ω, t) = ^c(t)

p(j , ω, t) ^{for j}=j^max(ω, t)and is 0 otherwise

Standard solution of the intertemporal maximization problem:

c·

c =r(t) −ρ

22/34

The unit elastic demand structure encourages the the highest possible price, while the existence of a competitive fringe sets a ceiling equal to the unit cost of the previous quality product

p^K(j^max(ω, t), ω, t) =λa^kw^K(t)

Monopoly profits accruing to global quality leaders from country K=W , E in sector ω will be equal to

π^K(ω, t) = (c^W(t) +c^E(t))N(t)

 1−¹

λ



For domestic subsidiary firms that produce in foreign country, the profits are

π^MNE(ω, t) = (c^W(t) +c^E(t))N(t)

 1− ^a

Mw^E(t) a^Ww^W(t)λ



23/34

Expected benefit of innovation equals the cost π^K(ω, t)

r(t) +I(ω, t) −

v.(ω,t) v(ω,t)

| {z }

v(ω,t)

I^K(ω, t) = (1−s^K)w^Kι^K

where subsidies s^K directly reduce the cost of R&D

Expected benefit of transfer, as the profit difference between the two production locations, equals the adaptive R&D cost

Arbitrage conditions

2κN

A^W (1−s^W)w^W = ^π

W

ρ+I^W−n, Western innovators

2κN

A^M (1−s^E)w^E = ^π

MNE

ρ+I^W+I^E−n− ^π

W

ρ+I^W−n, Production transfer 2κN

A^E (1−s^E)w^E = ^π

E

ρ+I^W+I^E−n, Eastern innovators

24/34

The share of sectors with Western leadership (ω^W), Eastern leadership (ω^E) and multinational production (ω^M) is stable in equilibrium and determined as

ω^W = ^I

W

I^M+I^W (1)

ω^E= ^I

M

I^M+I^W I^E

I^E+I^W (2)

ω^M= ^I

M

I^M+I^W I^W

I^E+I^W (3)

labor market clearing resource constraints

25/34

Welfare

W^K≡lnc^K

P + ^g

ρ−n where expenditure c^K is given by

c^KL^K= w^KL^K

| {z }

wage income

+

∑

k

π^kω^k

| {z }

profits

−

∑

k

w^kX A^kI^k

| {z }

R&D investment

and g=I^W+ (ω^M+ωE)I^E ln λ is theglobal growth rate

We denote(ω^M+ωE)as the competitive sectors

26/34

0 0.1 0.2 0.3 0.4 0.11

0.115

se

Iw

0 0.1 0.2 0.3 0.4

0.01 0.02

se

Ie

0 0.1 0.2 0.3 0.4

0.52 0.53 0.54

se

omega

Competitive sectors

0 0.1 0.2 0.3 0.4

0.7 0.8 0.9 1

se

we

Relative wage

0 0.1 0.2 0.3 0.4

3 3.2 3.4

se

g

Growth (%)

0 0.1 0.2 0.3 0.4

0 0.5 1

se

wel

Welfare, Eastern −−

i. Reduces Western profits and income - this is the BSE .

ii. Reduces Western innovation and production transfer. For high s^E, the transfer increases due to a fall in I^Eand adaptive R&D cost which compensates for the increase in relative wage

iii. Increases the growth rate of the global economy - this is the GRE .

iv. The effect on Western welfare depends on the strength of the GRE and BSE . _27/34

0 0.1 0.2 0.3 0.4 0.05

0.1 0.15 0.2

sw

Iw

Western innovation

0 0.1 0.2 0.3 0.4

0 0.01 0.02 0.03

sw

Ie

Eastern innovation

0 0.1 0.2 0.3 0.4

0.52 0.54 0.56

sw

omega

Competitive sectors

0 0.1 0.2 0.3 0.4

0.4 0.6 0.8

sw

wf

Relative wage

0 0.1 0.2 0.3 0.4

2.5 3 3.5 4

sw

g

Growth (%)

0 0.1 0.2 0.3 0.4

0 0.5 1

sw

wel

Welfare, Eastern −−

i. Reduces Eastern profits and income - this is the BSE .

ii. Reduces innovation in the East, but increases the presence of subsidiary firms.

Profits shift to West and Eastern relative wage falls.

iii. As long as the increase in Western innovation is strong enough, it increases the growth rate of the global economy - this is the GRE .

iv. The effect on Eastern welfare depends on the strength of the GRE and BSE . _28/34

s_n^W(s_n^E) =arg max W^W(s^W_n , s_n^E) (4)

s_n^E(s_n^W) =arg max n

W^E(s_n^W, s_n^E)^o (5) Increases in other country’s R&D subsidy trigger a defensive R&D subsidy response:

higher levels of Eastern subsidy lead to higher optimal strategic subsidy in the West s_n^∗W(s_n^∗E)and v.v.

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Sf

Sd

Best response subsidies

(Sd*,Sf*)=(0.273,0.198) best response Sd

best response Sf

Figure : Best response subsidies ^29/34

W^K≡ (ρ−n)U=lnc^K

P + ^g

ρ−n=Y^K−R^K+G , for K=W , E , (6)

∂W^W

∂s^W =^∂(R^W,Π^W)

∂s^W

| {z }

RCE (−)

+ ^∂G

∂s^W

| {z } GRE

(+) + ^∂Π

W

∂s^W

| {z } IBSE

(+)

+ ^∂Π

W

∂s^W

| {z } DBSE

(−)

, (7)

GRE - Consumers benefit from a higher-quality product, and also after it has been replaced by the next that builds on the previous quality ladder

DBSE - When a home firm innovates it drives another home firm out of business IBSE - Domestic innovation drives foreign firms out of business and shifts profits toward the home country

RCE - More resources must be allocated to R&D; as consumption is reduced, incumbent firms profits fall. However, higher DBSE increases production transfer which raises domestic profits and provide more labor for R&D - This mitigates the negative RCE

30/34

s^W_co(s_co^E; β) =arg maxn

T^W(s_co^W, s_co^E; β)^o

s_co^E(s_co^W; β) =arg maxn

T^E(s_co^W, s_co^E; β)^o where

T^K= (1−β)W^K+βW^W

∂T^W

∂s^W = ^∂R

W

∂s^W

| {z } RCE (−)

+ (1+β)^∂G

∂s^W

| {z }

GRE (+)

+ ^∂Π

W

∂s^W

| {z } IBSE

(+)

+β∂Π^E

∂s^W

| {z } IBSE

(−) .

Positive global gains from cooperation - (i) Neither government takes into account the GRE of its subsidies on the other country’s welfare and (ii) competitive subsidies do not take into account the negative IBSE on the other country

31/34

β s^W s^E W^W W^E W^total growth 0.00 0.36 0.4 0.489 0.319 0.808 3.80 0.25 0.32 0.4 0.475 0.337 0.812 3.69 0.50 0.28 0.4 0.457 0.358 0.815 3.58 0.75 0.24 0.38 0.460 0.352 0.812 3.46 1.00 0.08 0.23 0.524 0.239 0.763 3.03

Table : The effect of cooperation

Low levels of cooperation imply high subsidies, strong FDI and weak threat from the Eastern innovation

Intermediate levels of cooperation impose reduction in the s^W, less FDI, increase in the Eastern innovation, and a reduction in Western innovation and profits High cooperation lowers both subsidies, FDI is stronger, and the Eastern threat is low; growth falls

Medium levels of cooperation optimal for the EU as a whole but not implementable due to the negative effect on the Western welfare

32/34

33/34

Welfare effects of international competition in the market for innovations, the costs and the benefits of cooperative and non-cooperative R&D subsidies.

The overall effect of subsidies on domestic welfare depends on the relative strength of the two counteracting forces: GRE and BSE

When governments play a strategic subsidy game, increases in foreign competition trigger a defensive innovation policy mechanism Cooperation is beneficial for the global economy

34/34