Available via license: CC BYNC 4.0
Content may be subject to copyright.
Estudios económicos N° 76, Enero  Junio 2020. 147195 147
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...ACTORES, CONTRATOS Y MECANISMOS DE PAGO: EL CASO DEL SISTEMA DE SALUD DE NEUQUEN
° Rocha, L. A, Cardenas, L. Q, Alves Reis, F., Araújo Silva, N. G., &, Soares De Almeida, C. A.
(2021). Ination and innovation value: how ination affects innovation and the value strategy across
rms, Estudios económicos, 38(76), pp. 147195.
* Universidade Federal Rural do SemiÁrido, Brazil. Email: leonardoandrocha@yahoo.com.br.
ORCID: https://orcid.org/0000000327770702.
** Universidade Federal Rural do SemiÁrido, Brazil. Email: lquerido@ufersa.edu.br. ORCID: https://
orcid.org/0000000279663311.
*** Universidade Federal Rural de Pernabuco, Brazil. Email: felipe.alves.reia@hotmail.com. ORCID:
https://orcid.org/0000000335785357.
¤ Universidade Federal Rural do SemiÁrido, Brazil. Email: pie@ufersa.edu.br. ORCID: https://
orcid.org/0000000233425975.
¤¤ Universidade Federal Rural do SemiÁrido, Brazil. Email: alano@ufersa.edu.br. ORCID: https://
orcid.org/0000000283502094.
Estudios económicos. Vol. XXXVIII (N.S.), N° 76, Enero Junio 2021. 147195
ISSN 0425368X (versión impresa) / ISSN (versión digital) 25251295
INFLATION AND INNOVATION VALUE: HOW
INFLATION AFFECTS INNOVATION AND THE
VALUE STRATEGY ACROSS FIRMS°
INFLAÇÃO E O VALOR DA INOVAÇÃO: COMO A INFLAÇÃO AFETA A
INOVAÇÃO E A ESTRATÉGIA DE VALOR DAS FIRMAS
Leonardo Andrade Rocha*
Leonardo Querido Cardenas**
Felipe Alves Reis***
Napie Galve Araújo Silva¤
Carlos Alano Soares De Almeida¤¤
enviado: 9 junio 2020 – aceptado: 22 octubre 2020
Abstract
This research analyzes the effects of ination on R&D investments and innovation
driven growth. For this, an innovationdriven growth model was built in which
rms invest own resources and resources from nancial institutions. Credit costs
depend on the interest rate charged by these institutions. In an inationtargeting
regime, the monetary authority adjusts the nominal interest rate in order to converge
current ination to the established target. It adjusts the interest rate of nancial ins
titutions, changing the opportunity cost of investments. As a result, rising ination
Estudios económicos N° 76, Enero  Junio 2020. 147195148
ESTUDIOS ECONOMICOS
promotes a reduction in R&D investments demand, reducing the rate of technolo
gical progress. In the empirical exercise of the model, the estimated coefcient of
elasticity of R&D investments is negatively affected by ination.
Keywords: Innovation, ination, R&D.
JEL Codes: E41, O41.
Resumo
O presente estudo analisa os efeitos da inação nos investimentos em P&D e no
crescimento orientado pela inovação. Para isto, foi construído um modelo de cresci
mento orientado pela inovação de forma que as rmas investem recursos próprios e
captam outros recursos junto às instituições nanceiras. Os custos dos empréstimos
dependem da taxa de juros cobrada pelas instituições nanceiras. Em um regime
de metas, a autoridade monetária ajusta a taxa de juros nominal no intuito de fazer
convergir a inação corrente para a meta estabelecida. Este processo ajusta a taxa
de juros cobrada pelas instituições nanceiras elevando o custo de oportunidade
dos investimentos. Como resultado, o aumento da inação promove uma redução
na demanda por investimentos em P&D, reduzindo a taxa de progresso tecnológico.
No exercício empírico do modelo, o coeciente estimado de elasticidade dos in
vestimentos em P&D é negativamente afetado pela inação.
Palabras chaves: inovação, inação, I&D.
Classicação JEL: E41, O41.
Estudios económicos N° 76, Enero  Junio 2020. 147195 149
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
I. INTRODUCTION
Economic theory has always pointed to the harmful effects of ination on
the growth of economies, whether through expectations, the costs of investing, the
difculty of predicting relative prices in the future or even the political aspects
associated with austerity measures of macroeconomic policy (Dressler, 2016). In
general, it is understood that macroeconomic policy is clearly important for eco
nomic growth because of its role in reducing uncertainty and encouraging invest
ment by economic agents (Barro & SalaIMartin, 2004; Aghion & Howitt, 2009;
Acemoglu, 2009; Ramzi & Viem, 2016).
While this broad relationship has long been debated over the years, few
studies have analyzed the transmission channels which ination exerts on specic
investments, including those that are fundamental to national economic progress:
investments in research and development (R&D). Since the contributions of Solow
(1956) on how technological advances are critical to the accumulation of wealth in
economies, many important studies sought to demonstrate the role of technology
and its development in the differences in growth between the various economies
(Hall & Jones, 1999; Aghion & Howitt, 2009; Acemoglu, 2009).
Although the debate about ination costs related to growth has been the subject
of several studies in the last 5060 years (Blanchard, 2016; Brunnermeier & Sannikov,
2016), few studies have examined the impact on specic investments in R&D and,
hence, in innovation. Such investments, as Aghion and Howitt (2009), Hall (2002),
Hall, Lotti, and Mairesse (2013), Hall, Mairesse, and Mohnen (2010), and Aghion,
Howitt, and Prantl (2015), represent one of the basic inputs for innovation and techno
logical progress, since it aligns the creation of a new technology, whether in products,
processes, or forms of management, with the rms’ value strategy (Coad, 2011).
In this perspective, these features have very particular characteristics, since
they represent over 50% of the wages for a highly qualied workforce. In this case,
human resources become a valuable asset of the rms and are subject to agreements
of unique characteristics (Hall, 2002). In this way, the predictability of assets and
the formation of prices are crucial in the decision making on investments in R&D,
as such prices reect a possible anticipation of the future growth of rms (Kung &
Schmid, 2015). In addition, recent empirical evidence shows that investments in
R&D are strongly affected by the cash requirements of enterprises.
Considering these aspects, the following questions arise: how does ination
affect the ‘innovation x market value’ of rms? Can ination represent an advantage
Estudios económicos N° 76, Enero  Junio 2020. 147195150
ESTUDIOS ECONOMICOS
for the rms’ value strategy? In order to answer these questions, this study developed
a Schumpeterian growth model, relating the efforts in innovation with the resource
constraints of rms. The biggest advantage of this theoretical approach is to study
the rms’ behavior in conditions of competition, where the prize for innovation is the
temporary monopoly associated with the creation of technology (Aghion, Akcigit, &
Howitt, 2013). As highlights of the theoretical model, we can list:
1. Firms have nancial resources constraints, so that part of the investments are
funded by nancial institutions through loans;
2. Credit costs are assigned by the interest rate charged by banks that depend on
the spread established plus the interest rate set by the monetary authority in
order to converge the current ination to the target set (targeting regime);
3. Unlike Chu and Lai (2013) and Chu et al. (2015), the inclusion of targets in the
model has to be a more realistic way to measure the effect of ination by adjus
ting the interest rate, given the popularity of the targeting regime (Wash, 2003;
Ozdemir & Tuzunturk, 2009; Umar, Dahalan, & Aziz, 2016; Hosny, 2017). In
this case, the target system becomes important in three factors: (1) in addition
to controlling ination, it reduces its volatility over time; (2) it minimizes the
real costs of disination; (3) it approaches the longterm ination expectations
established by the target (Capistran & RamosFrance, 2010).
The model results indicate a negative effect between ination on the demand
for R&D and, hence, the rate of technological progress. In the empirical exercise
model, the results reveal that ination reduces the elasticity coefcient of R&D in the
market value of rms. Moreover, a nonlinear relation is identied between ination
and the value of rms; low and moderate levels have a positive relationship with the
value, while high levels imply reduction. The study’s ndings indicate that modest
ination may have a positive alignment with the rms’ value strategy, but with a
negative result on the investment in applied R&D (elasticity coefcient).
II. THEORETICAL MODEL
Many models disregard the inuences of the nancial market and nancial
intermediation (banks) in the growth process and how this process is related to
efforts in innovation by businesses. This is because rms always seek to nance
part of the total investments allocated to R&D activities (Acemoglu, 2009).
An important and recent contribution in this regard is found in Chu et al.
(2020), who dealt with the role of credit restrictions in stimulating or not stimula
Estudios económicos N° 76, Enero  Junio 2020. 147195 151
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
ting innovation, according to different patent protection regimes. Thus, in markets
with credit restrictions, strengthening patent protection implies limiting/stiing
local R&D demand, compromising innovation.
However, each economy has a relatively distinct economic environment,
especially because different types of monetary policies are applied to ensure price
stability. The traditional monetary policy implemented in several countries is orien
ted towards controlling the economy’s basic interest rate. In this way, the central
bank inuences the loan rates that are offered by banks, both nonnancial institu
tions and individuals, conditioning, hence, the dynamics of ination and economic
growth (Becker, Orborn, & Yildirim, 2012).
II.2. A Schumpeterian Economy
Growth models based on Schumpeterian assumptions have attracted the atten
tion of many researchers since they have highlighted the key role of innovation for
economic growth. This growth is due to the development of innovations that lead to
the “destruction” of current technologies, making them obsolete, and replacing them
with a new generation of techniques and products (Aghion & Howitt, 2009).
The model is based on a discrete sequence of time periods . In each period
there is a stock of labor consisting of L individuals who work aimed at maximizing
the expected consumption and that, in this study, will be normalized to a unit (L
= 1). This normalization follows the approach presented by Aghion and Howitt
(2009), as a way of reducing the model, without many losses of generality, when
considering that individuals, each of whom, live only for that period t. Thus, we
restrict the effects of population growth to the model.
The economy has a xed population L, which we normalize to unity.
Everyone is endowed with one unit of labor services in the rst period and
none in the second, and is risk neutral. (Aghion & Howitt, 2009, p. 130).
The nal product is created using a ux “i” of intermediate inputs ( )
continuous under the condition , according to the production function:
(1)
The productivity parameter reects the quality of the intermediate input of sector
“i” in the time “t”. Although production of the nal good will occur in a competitive
Estudios económicos N° 76, Enero  Junio 2020. 147195152
ESTUDIOS ECONOMICOS
market, the intermediate inputs sector is monopolized by the leading rm in the
current technology. In this case, the monopolist enjoys prots in the short term, as
they create the new current generation of inputs with the best quality. The demand
curve for the monopolist is given by the partial derivative:
(2)
The monopolist seeks to maximize the profit function of the sector
(), replacing the demand curve in function:
(3)
The monopolist’s equilibrium prots are obtained by replacing (3) on the prot
function, giving the equilibrium level:
(4)
II.3. R&D and Technical Progress
Advances in productivity occur as improvements in future generations of
inputs, so that each new generation implies a signicant advance in the current quality:
(5)
In certain situations, the innovation carried out does not achieve the expec
ted result and the improvement is not accepted in the market. In this case, producti
vity does not increase and is assumed to remain unchanged, . The size
of innovation is given by the parameter, exogenously determined. To increase
the chances of a successful innovation, the entrepreneur funds research activity
through large investments in R&D, which is represented by the variable . Thus,
the greater the effort in innovation through applying considerable resources, the
greater the chances of success for the research and, therefore, for the new tech
nology. The function that captures the probability of innovation success is called
‘innovation function’:
Estudios económicos N° 76, Enero  Junio 2020. 147195 153
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
(6)
According to Equation (6), the parameters represent, respectively, the
research productivity and the elasticity coefcient of R&D. In this sense, the more
the productivity can advance technologically in the next period, the greater the
probability of success of the research:
(7)
Technological advances observed in the sector are the residual advance of
expected productivity, since investments in research are always subject to uncertainty:
(8)
If innovation is successful, prots in the industry are appropriated by the
monopolist. However, in the absence of success, the entrepreneur has the sunk
costs equivalent to the total investment. Expected prots are adjusted for produc
tivity, reecting that the entrepreneur seek the highest prots by generated assets
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚଵ݈݃(ܴ&ܦ)௧ +ߚଶ݂݈݅݊௧ +ߚଷ[݈݃(ܴ&ܦ)כ ݂݈݅݊]௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
.
Thus, the entrepreneur allocates the R&D resources to maximize expected
prots adjusted for productivity:
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚଵ݈݃(ܴ&ܦ)௧ +ߚଶ݂݈݅݊௧ +ߚଷ[݈݃(ܴ&ܦ)כ ݂݈݅݊]௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
(9a)
(9b)
(9c)
The rate of technological progress is obtained by substituting (9c) to (8):
(10)
Estudios económicos N° 76, Enero  Junio 2020. 147195154
ESTUDIOS ECONOMICOS
II.3. BankFunded Researches
Only a portion of the investment in research is funded with the monopolist
entrepreneur’s own resources. Another part is acquired through loans with banks.
Take entrepreneur’s income as . Banks nance only a portion of the investment,
forcing the entrepreneur to have nancial guarantees. Assuming the monopolist
allocates all their income, the nancing they get is the difference equivalent to
. Banks, on the other hand, charge interest (r) on the total invest
ment to offset losses generated by funded projects, which subsequently did not have
the desired economic viability. The interest rate is based on the following formula:
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚଵ݈݃(ܴ&ܦ)௧ +ߚଶ݂݈݅݊௧ +ߚଷ[݈݃(ܴ&ܦ)כ ݂݈݅݊]௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
(11)
In (11),
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚଵ݈݃(ܴ&ܦ)௧ +ߚଶ݂݈݅݊௧ +ߚଷ[݈݃(ܴ&ܦ)כ ݂݈݅݊]௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
represents the interest rate determined by the monetary autho
rity, and is the additional costs that set the bank interest. The basic interest rate
of the economy adjusts the nal interest charged by banks on loans, serving as a
‘minimum’ for the denition of money opportunity cost. The high risk involved in
nancing activities raises an additional cost which, together with the absence of
limiting mechanisms in tariff charges, adds the composition of the bank spread1.
Sunk costs of research are now represented by the total volume invested plus
interest charged as a result of nancing. The sum of the two components denes
the total cost of the research: .
In this scenario, we highlight an important limitation of the theoretical
model: secondary effects of ination on important measures of rms. Kang and
Pueger (2015, pp.115117) found evidence that the effects of ination can help
explain, at least, a direct variation in credit spreads, in addition to volatility in
stocks (e.g. Aliyu (2012)) and in the dividend indexprice. As Kang and Pueger
(2015) argued:
We nd that ination risk can explain at least as much variation in credit
spreads as can equity volatility and the dividendprice ratio. First, more
volatile ination increases the ex ante probability that rms will default
due to high real liabilities. Second, when ination and real cash ows
1 Following Gropp, Sørensen, and Lichtenberger (2007), an important difference between lending
rates and market interest rates can be attributed to credit risk, reecting the likely possibility that
some loans may not be fully paid by agents. See Were and Wambua (2014).
Estudios económicos N° 76, Enero  Junio 2020. 147195 155
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
are highly correlated, there is a risk of low ination recessions. In this
case, low real cash ows and high real liabilities tend to hit rms at the
same time, and this interaction increases default rates and real investor
losses. Moreover, ination cyclicality may also increase the default
risk premium in credit spreads if investors are risk averse. (Kang &
Pueger, 2015, pp. 115116).
Such factors contribute, to some extent, to a nal effect on the value of rms
in the stock market. Other evidence also points to the reexes on the behavior of
companies, which adjust their capital structure in response to the risk of persistent
ination over time (e.g. Hackbarth, Miao, & Morellec, 2006; Chen, CollinDu
fresne, & Goldstein, 2009; Bhamra, Kuehn, & Strebulaev, 2010; Gomes & Schmid,
2010; Gourio, 2013).
Although this model has been limited in the consequences of ination in
interest rates, an improvement of the model in the future may contribute to greater
theoretical robustness.
Reframing (9.a), the new entrepreneur optimization problem incorporates
the total cost of the research:
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚଵ݈݃(ܴ&ܦ)௧ +ߚଶ݂݈݅݊௧ +ߚଷ[݈݃(ܴ&ܦ)כ ݂݈݅݊]௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
(12a)
(12b)
(12c)
As in Equations (12b) and (12c), the interest rate acts discounting the effective
value of the investment, reecting a smaller amount of allocated resources in R&D
activities and, consequently, in the success of innovation (measured by probability).
Substituting (12c) into (8), the equilibrium technical progress rate is obtained:
(13)
Estudios económicos N° 76, Enero  Junio 2020. 147195156
ESTUDIOS ECONOMICOS
(14)
Equation (14) illustrates the inuence of interest rates on the rate of tech
nological progress, negatively affecting it. Thus, in economies with higher interest
rates, the demand for investment tends to be lower, reducing the necessary efforts
to sustain the rate of technological progress.
II.4. Introducing Monetary Policy
The role of the central bank, either in developed or developing economies,
is focused on the pursuit of price stability, making the basic interest rate one of
the main instruments of monetary policy (Stein, 2012). Several authors argue that
the choice of a price index, such as monitoring over time, was gradually guided
by the idea that ination is, in fact, a monetary phenomenon (Goodfriend, 2007;
Mishkin, 2007, 2008; Wynne, 2008; Stein, 2012; Anand, Prasad, & Zhang, 2015).
As observed by Taylor (2000, p. 90), “monetarypolicy decisions are best thought
of as rules, or reaction functions, in which the shortterm nominal interest rate (the
instrument of policy) is adjusted in reaction to economic events.”
This model assumes that the monetary authority acts by controlling the
monetary policy in order to converge the current ination toward the center of the
target set. Thus, the deviation caused between current ination and the center of
the target determines the position of the authority to increase or reduce the interest
according to the direction of the deviation. This relationship can be expressed as
the change in the interest rate (r *):
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚଵ݈݃(ܴ&ܦ)௧ +ߚଶ݂݈݅݊௧ +ߚଷ[݈݃(ܴ&ܦ)כ ݂݈݅݊]௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
(15)
According to Equation (15), ination ( ) is measured by the percentage
change in the general price level, integrating all sectors of the economy. Thus, the
monetary authority establishes a target ( ), in order to adjust the interest rate
to the extent that the current ination deviates from the established target. When
ination exceeds the target, the central bank increases the rate of interest in order
to ‘level’ economic activity, hence converging ination to the target set. The para
meters , represent, respectively, the shift term and the elasticity of the ination
deviation.
Estudios económicos N° 76, Enero  Junio 2020. 147195 157
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
II.5. Consequences of Ination Rates Persistently Above Target
In economies where ination runs persistently above the target set by the
monetary authority, it is common to observe high interest rates. As a result, the
demand for investments decreases, especially in R&D activities, whose return on
investment consists of medium and long term (Hall, Lotti, & Mairesse, 2013).
Modifying Equation (12b) incorporating (15), we have:
(16)
As Equation (16), a persistent rise in ination above the target prompts
the monetary authority to increase interest rates and, consequently, the demand
for investments in R&D decreases. With higher interest rates, the cost of capital
rises, reducing demand and efforts in innovation. This reduction, in turn, implies a
lower probability of success of the innovative entrepreneur, restricting the rate of
technical progress in the industry.
(17)
By integrating the sectors of the economy, the rate of technical progress in
this economy is reached:
(18)
According to Equation (18), we can observe an inverse relation between
ination and the rate of technological progress in the economy. This relation is not
as recent as it may seem, although few studies have focused their analysis on this
subject. A pioneering and major study linking the effects of ination on innovation
consists of the contributions of Manseld (1980). According to the author, as ination
reduces investment rates, it discourages the demand for machinery and equipment, as
well as expansion of new plants, limiting the application of specic investments such
as R&D. Another important ination effect consists of public research funding. In
developing economies, most of the investments in R&D activities come from public
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚଵ݈݃(ܴ&ܦ)௧ +ߚଶ݂݈݅݊௧ +ߚଷ[݈݃(ܴ&ܦ)כ ݂݈݅݊]௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
Estudios económicos N° 76, Enero  Junio 2020. 147195158
ESTUDIOS ECONOMICOS
funds. In these circumstances, governments may be compelled to ‘cut’ a portion of
the budget as part of the goal of antiinationary scal policy.
In this way, research funding may be limited in the implementation of future
budgets, restricting the longterm productivity growth. Thus, in order to control
rising ination, economic policy can be directed to promote an ‘undesirable effect’
to reduce the longterm growth2. Such institutional efforts can explain how deve
loping economies with higher ination rates also have signicant limitations for
convergence towards the technology frontier. Thus, in the presence of credit cons
traints, through funding research, ination has a direct impact on the nancing
interest, raising the opportunity cost of investments in R&D (Chu & Lai, 2013).
III. RESEARCH METHODOLOGY
III.1. Data Source and Sample Delimitation
The data used in this study were extracted from an important data source:
Standard and Poor’s (S&P) Capital IQ Platform. The S&P Capital IQ Platform is
an important source of nancial information, containing nancial data from more
than 1 million rms worldwide. The main advantage of this platform is the brea
dth of data, separated by countries and sectors, in addition to including the most
important nancial indicators of companies, which allows a more detailed analysis
of the rms’ strategy. The following lters were used:
(i) active rms in the world with legal origin by country and dened market value
(publicly traded companies);
(ii) Firms classied according to international credit rating industry classication3;
(iii) Firms identied with the digit1 of the ‘SIC Codes’ corresponding to 10 sectors;
(iv) Financial variables from 2010 to 2015 (six years);
(v) The nal sample was 34 194 rms from 125 countries distributed over the
years 2010 to 2015, corresponding to a panel with 205 164 observations.
2 “Serious ination can have a signicant effect on government nanced R&D if it stimulates an
antiinationary tax policy that affects the size and type of government R&D programs” (Manseld,
1980, p. 1093).
3 The Standard Industrial Classication (SIC) offers 4digit industry classication (Sector, majority group,
group of industries, and industry). For more details, see: http://siccode.com/en/pages/whatisasiccode
Estudios económicos N° 76, Enero  Junio 2020. 147195 159
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Macroeconomic data were extracted from a relevant international rating
company, which calculates risk indicators for 140 countries in the world, Political
Risk Services Group. This company measures and analyzes twentytwo variables
to dene and estimate risk prediction models for international investors divided into
three subgroups in the International Country Risk Guide report: twelve political risk
variables, ve nancial risk variables, and ve economic risk variables. An impor
tant advantage of the available database consists of a wide time period from 1984
to the present, which allows a better control on the forecast measures and adequacy
of variables (Charron, 2011). Given the importance of the quality of risk measures,
recent studies have applied the basis of risk analysis at the macroeconomic level
(Osabutey & Okoro, 2015; Stockemer, 2013; Kunieda, Okada, & Shibata, 2016;
Myles & Youse, 2015; Beal & Graham, 2014).
III.2. Operation and Denition of Variables
The variables used in the study are presented in Table 1, with a distinction
between variable groups by micro and macro level:
Table 1. Description of the model variables
Variables Denition Applied research
Financial variables (microlevel)
R&D
Investments made in research activities
and which may occur internally in rms or
externally, through universities and research
institutes. Such investments represent the
nancial effort of the rm in order to nance:
new product development on innovative
technology formulation, process development,
and processes performed in product update or
existing service line.
Hall, Mairesse, & Mohnen
(2010), Bogliacino &
Cordona (2010), Hall,
Lotti, & Mairesse (2013),
Montresor & Vezzani
(2015), and Kancs &
Siliverstovs (2016)
mkt_cap Market capitalization value. Dias (2013)
capex
Investments in order to acquire or upgrade
physical assets such as equipment, properties,
and industrial plants.
Hall (2002), Hall, Lotti,
& Mairesse (2008, 2013),
Hall, Mairesse & Mohnen
(2010), Gupta, Banerjee, &
Onur (2017)
Estudios económicos N° 76, Enero  Junio 2020. 147195160
ESTUDIOS ECONOMICOS
ATV Total assets. Hall (2002) & Hall, Lotti,
and Mairesse (2008, 2013)
QTobin Ratio of market capitalization to total assets.
Coad (2011), Gupta, Onur,
& Banerjee (2017), and
Hall, Jaffe, & Trajtenberg
(2005)
LT_inv Longterm investments represent investments
held for more than a year in the rm’s activities.
Graham, Campbell &
Rajgopal (2005), Lerner,
Sorensen, & Strömberg
(2011), and Bourke &
Roper (2017)
ST_inv
Shortterm investments represent relatively
liquid investments, i.e. activities of the rm of
more than three months and less than one year.
Bourke & Roper (2017)
and Cremers, Pareek, &
Sautner (2017)
Macroeconomic variables
gdp_
growth
GDP growth rate, calculated as the percentage
change of the current year compared to the
previous year.
Bashir (2002), Aghion &
Howitt (2009), and Aghion
& Jaravel (2015)
gdp_
budget
Corresponds to the balance of the central
government budget (including grants) for
a given year in the national currency and is
expressed as a percentage of GDP this year in
the national currency.
Aghion & Marinescu
(2007), Aghion, Hémous,
& Kharroubi (2014)
gdp_
current
Corresponds to the balance of payments
balance for a given year, converted into US
dollars at the average exchange rate for that
year. It is expressed as a percentage of GDP,
converted into US dollars at the average
exchange rate for that year.
Gehringer (2013) and Ege
& Ege (2017)
gdp_exp
Corresponds to the balance of payments
balance for a given year, converted into US
dollars at the average exchange rate for that
year. It is expressed as a percentage of total
exports of the country’s goods and services,
converted into US dollars at the average
exchange rate for that year.
Aghion & Marinescu
(2007) and Gehringer
(2013)
Estudios económicos N° 76, Enero  Junio 2020. 147195 161
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
in
Ination rate that is obtained annually through
the unweighted average of the Consumer
Price Index.
Bashir (2002), Funk &
Kromen (2010), Chu & Lai
(2013), Ascari & Sbordone
(2014) and Anand, Prasad
& Zhang (2015)
Source: Prepared by the authors.
Note: The column “Applied Research” lists the studies that used the variables cited in research related
to the topic discussed.
III.3. Regression Model
To analyze the effects of investments in R&D on rms’ innovation, a portion
of the literature on the topic applies traditional sales performance measures to asso
ciate with investments in research. However, we prefer to adopt the value of rms
as a performance measure, based on Tobin’s Q indicator. This choice minimizes
potential problems related to time lags between the rm’s behavior and changes
in performance, since “future performance gains obtained through ‘appropriate’
behavior can be anticipated on the stock market and can thus be included into a
rm’s current market value (and hence Tobin’s q)” (COAD, 2011, p. 1054).
Moreover, the market value better reects the returns of innovation, in addi
tion to allowing a better comparison of the productivity of innovation according to
different markets and production technologies, making it difcult to compare the
productivity levels of these rms (Hall, Jaffe, & Trajtenberg, 2005; Hall, Mairesse,
& Mohnen, 2010).
To study the effect of ination in relation to “investment x performance”,
the following equation was estimated:
Eq. 1
1. Page 153:
Expected profits are adjusted for productivity, reflecting that the entrepreneur
seek the highest profits by generated assets (ȫ
௧
כ=ȫ
௧ ܣ௧
Τ)
Equation (9a): ܴ
௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെ ܴ௧ൟ
2. Page 154:
Equation (11): ݎ=ݎכ+ܵ
In (11), ݎכrepresents (…)
3. Page 155:
Equation (12a): ܴ
෨௧ =ܽݎ݃݉áݔ൛ߤ௧ȫ
௧
כെܿ(ܴ௧)ൟ
4. Page 156:
Equation (15): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
5. Page 157:
Equation (18): ݎכ=߭+߶(ߨ௧െ ߨ) ߨ௧=݈݊ ቀ
షభቁ; ௧=௧݀݅
ଵ
6. Page 161:
Eq.1
݈݃(்ܳ)௧
=ߙ+ߚ
ଵ
݈݃(ܴ&ܦ)
௧
+ߚ
ଶ
݂݈݅݊
௧
+ߚ
ଷ
[݈݃(ܴ&ܦ)כ ݂݈݅݊]
௧
+ߚସ(݂݈݅݊)௧ଶ+܆ࢽ+܆ࢾ+ߤ+ߨ+߬௧+ߝ௧
7. Page 179:
Second paragraph: Replace "Figures 2 and 3 illustrates" for "Figures 2 and 3
illustrate"
According to Eq.1, the rm’s value, calculated through the QTobin, is regres
sed with R&D, ination, crosseffects between ination and investment and ination
rate squared. The Xmicro vector represents the set of variables at the nancial level
Estudios económicos N° 76, Enero  Junio 2020. 147195162
ESTUDIOS ECONOMICOS
of rms that help to control important characteristics of the value strategy (capex,
assets, and short and longterm investments). The other vector, Xmacro, relates to
the macrolevel dimensions that condition rms’ strategies and help to control the
greater latent effects of ination on economies. The subscripts i, j, r, and t represent,
respectively, the dimensions at the level of rm, sector, economic region, and year.
Additional vectors , , correspond to the xed effects at the sectoral,
regional, and temporal level, affect the rms’ value strategy based on the invest
ments made and control the effects at the macroeconomic level in the ‘investment
x performance’ relation. The stochastic error is captured , representing all
other factors that are not part of the research scope, being irrelevant to the model.
III.4. Estimation Method and Robustness
Eq.1 can be estimated by the traditional ordinary least squares technique
(OLS), grouping the data and disregarding the xed effects in the main model.
However, the absence of xed effects may lead to a serious bias in estimates that do
not disappear even when the sample is relatively large (Greene, 2012). In this case,
regional and sectoral factors exert a signicant inuence, either on the demand for
investments or on the macroeconomic scenario, through differences between scal
and monetary policies, affecting ination rates. Specic market structures can induce
specic investment demands, affecting the relation between regressors and the secto
ral xed effects (Coad, 2011). Such movements can present a systematic correlation
with the stochastic disturbance, leading to an inconsistency in the estimates.
The inclusion of xed effects controls such latent effects of the stochastic
disturbance and the covariates. In this case, the regression technique with traditio
nal panel data includes another important factor in the size of the xed effects: the
individual factor or individual heterogeneity effect. However, the sample is cha
racterized by an important singularity that consists of a panel with a crosssectional
dimension much larger than the temporal section (called short panel).
In a sample with these characteristics, recent research, especially Hahn and
Kuersteiner (2002), Hahn and Newey (2004), Lee and Phillips (2015), and Bester
and Hansen (2016), have pointed out the relative problems in models with panel
data when the crosssection is considerably higher than the temporal cut (short
panel) – see Hsiao (2014). This phenomenon, called “incidence of parameters pro
blem”, was rstly diagnosed by Neyman and Scott (1948).
Estudios económicos N° 76, Enero  Junio 2020. 147195 163
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
As shown by Hahn and Newey (2004), the parameters of the variables of
interest are inconsistent when the number of individuals (n) becomes sufciently
large relative to the time period (T). This inconsistency stems from the nite num
ber of observations that are available to estimate each individual effect (signi
cantly reducing the degree of freedom of the model).
An alternative form, proposed by Bester and Hansen (2016), consists in esti
mating the xed effects by aggregating the individual effects at different levels of
predetermined groups. In the same way that the authors in the study, in microdata
on teaching evaluation, students can be grouped according to classes, education
levels, schools, and districts; or even rms can be grouped according to different
sectoral levels or specic economic regions. This technique, called ‘grouped effects
estimator’, considers the estimation of model parameters by treating individual
specic effects as constant within groups at a particular level.
Therefore, this research implements an adaptation of the model proposed by
Bester and Hansen (2016) in nonlinear functions for the panel data methodology.
In this case, the executed technique excludes the individual effects, aggregating
them by regional and sectorial level, executing a procedure of grouped xed effects.
This derives from a large amplitude of the crosssectional sample that includes
34,194 rms over 6 years. The inclusion of xed effects at the rm level would
entail serious damage to the model, what makes necessary treatment from the tech
nique proposed by Bester and Hansen (2016) adapted to a linear model. Lastly, the
parameter covariance matrix was estimated using the residual clustering technique,
having the individual units as dimension to correct serial autocorrelation and hete
roscedasticity. Thus, the standarderror estimative are consistent and parameters
are efcient (Greene, 2012).
Although, in the model, a more indepth investigation is not applied on the
existence or not of stationarity in the data, the presence of panel data with very
large crosssection units (and Tsmall) make these limitations relatively easier to
be addressed. Circumvented, assuming some restrictions, such as homogeneity
in the slope coefcients (e.g. Baltagi, 2005, pp. 201) and independence of obser
vations between the crosssectional units (e.g. Hsiao, 2014, p. 9). Although this
assumption is strong, the control of heterogeneity to the model can be sustained
in many situations. These advantages are relatively greater in panel models with
Nlarge, which are adherent with the sample of the present study. In addition, the
comparison with the dynamic panel technique allowed a better correlation of the
results that demonstrated a relative convergence in the understanding. Furthermore,
based on the assumption that the observations between the transversal units are
Estudios económicos N° 76, Enero  Junio 2020. 147195164
ESTUDIOS ECONOMICOS
independent, especially in samples with Nlarge and Tsmall (e.g. Hsiao, 2003,
p. 7; 2007, p. 5; 2014, p. 386387), we can appropriate the Central Limit Theorem
between the transversal units to show that the distributions of many estimators
remain asymptotically normal (e.g. Binder, Hsiao, & Pesaran, 2005, Im, Pesaran,
& Shin, 2003), Levin, Lin, & Chu, 2002). It is also necessary to point out that, in
series at the level of microdata, relatively large crosssectional units with limited
time space are common, involving relatively small T values, so it is natural to
assume that the series follow stationary processes (see Hall & Urga, 2000, p. 2).
III.5. Dynamic Effects of Firms’ Value
Another way to estimate the Eq.1 model is to include the lagged effect of
the dependent variable. In general notation, this implies adjusting the model as:
Eq.2
The parameter to be estimated “ ” captures the persistent effect of the
rm’s value over time. The vector of parameters of the independent variables is
determined by “ ”. The xed effects related to the rm level are captured by “ ”,
and the temporal effects by “ ”. The error term is dened by “ ”.
One of the great contributions in the econometrics of dynamic models con
sists of the study developed by Arellano and Bond (1991), who proposed the use of
the Difference GMM estimator (Generalized Method of Moments). This procedure
consists in transforming the data through differences in time and addressing the
problem of endogeneity through the use of lagged values as instruments. Subse
quently, this technique demonstrated limited performance, especially in conditions
close to the present study: when the time cut is relatively small compared to the
crosssection (crosssection, N> T) and when the dependent variable tends to show
a persistence pattern in time. Such factors are limiting to the difference GMM
technique, being subject to a large sample bias (Arellano & Bover, 1995; Blundell
& Bond, 1998; AlonsoBorrego & Arellano, 1999; Arellano, 2016; Jha, 2019).
Based on the contributions of Arellano and Bover (1995) and Blundell
and Bond (1998), the System GMM estimator sought to solve such problems:
(1) increasing efciency, since it uses more moment conditions than the difference
GMM, which makes it more appropriate for nonstationary data and; (2) ensuring
consistency, since it does not depend on the assumption of any secondorder serial
Estudios económicos N° 76, Enero  Junio 2020. 147195 165
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
correlation (Mehic, 2018). In addition, the combination of factors such as short
panel (N> T), lagged dependent variable, inclusion of numerous xed effects and
a lack of good external instruments, make the technique even more attractive in
empirical studies (Roodman, 2009).
One of the great advances in the technique is to limit the inclusion of ins
truments to the model, avoiding the “proliferation of instruments”. In this critical
problem, the results of the model may suggest a validity, when, in fact, the model
is closer to being invalid. This is known in the literature as a false positive, leading
to conclusions that are precipitated by the excessive inclusion of nonrelevant ins
truments, inating the model (Roodman, 2009).
The solution to this problem is to “collapse” the instrument matrix, limiting
the entry of many outdated instruments. This procedure is duly presented in the
contributions of Beck and Levine (2005), Carkovic and Levine (2005) and, later,
in Roodman (2009) and Labra and Torrecillas (2018).
IV. RESEARCH RESULTS
IV.1. Descriptive Analysis of the Sample
Table 2 shows the distribution of rms according to the seven geographical
regions in the world.
Table 2. Distribution of rms by geographic region
Geographic location Freq. Abs. Freq. Rel. (%) Freq. Cum. (%)
Africa / Middle East 1 645 4.81 4.81
Asia / Pacic 14 494 42.39 47.20
Caribbean 201 0.59 47.79
Central America and Mexico 91 0.27 48.05
Europe 5 976 17.48 65.53
Latin America and Caribbean 448 1.31 66.84
United States and Canada 11 339 33.16 100.00
Total 34 194 100.00 
Source: prepared by the authors
Estudios económicos N° 76, Enero  Junio 2020. 147195166
ESTUDIOS ECONOMICOS
As shown in the table, most of the rms in the sample are in Asia, which
represents approximately 42% of the total. Countries from North America (United
States of America and Canada) represent approximately 33% and Europe 17.5%.
Latin America and the Caribbean correspond to 448 rms, representing about
1.31% of the sample. Africa and the Middle East amount to 1 645 rms (4.81%
of the sample).
The distribution of rms by sector, as SIC Codes classication, is shown in
the results of Table 3.
Table 3. Distribution of rms by sector
Sector Freq. Abs. Freq. Rel. (%) Freq. Cum. (%)
Division A: Agriculture,
Forestry, and Fishing 205 0.60 0.60
Division B: Mining 3 171 9.27 9.87
Division C: Construction 679 1.99 11.86
Division D: Manufacturing 12 709 37.17 49.03
Division E: Transportation,
Communications, Electric, Gas,
and Sanitary Services
2 133 6.24 55.26
Division F: Wholesale Trade 1 083 3.17 58.43
Division G: Retail Trade 1 174 3.43 61.86
Division H: Finance, Insurance,
and Real Estate 8 486 24.82 86.68
Division I: Services 4 177 12.22 98.90
Division J: Public
Administration 377 1.10 100.00
Total 34 194 100.00 
Source: Authors’ calculations.
Estudios económicos N° 76, Enero  Junio 2020. 147195 167
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
The manufacturing sector comprises the largest share of rms in the sample,
representing a total of 12 709 rms and a relative value of 37.17%. The nancial
sector (8 486 rms) accounts for 24.82% of the total. Next, we have the service
industries and transport, communications and associates, totaling respectively 4 177
rms (12.22% of the sample) and 2 133 rms (6.24% of the sample). Public admi
nistration rms include 377 rms, what is about 1% of the total sample. Lastly, the
agricultural activity has only 205 rms (0.6% of the total sample).
Table 4 shows the distribution of rms across geographic location and the
sector. Crossdistribution allows a better monitoring of the business activity by
economic region.
In Asia, most rms are distributed in the nancial and service sectors. These
industries comprise a total of 3 823 rms, representing approximately 26% of the
total from the region. The USA and Canada have a higher number of rms in the
mining, manufacturing, nancial and services, covering a total of 10 012 rms,
almost 90% of this region.
In Europe, manufacturing and nancial sectors correspond to 3 458 rms,
totaling a relative value of 58% of this region. In Latin America and the Caribbean,
manufacturing and nancial sectors concentrate most of the rms, with a total
of 259 companies that correspond to a relative value of approximately 58% of
this region. In regions of Africa and the Middle East, the predominant sectors are
manufacturing, transportation, communications, electric, gas and sanitary services,
nancial and services. Together, these sectors covered 1 367 rms, what is appro
ximately 87% of the regional sample.
Table 5 shows the correlation matrix between the model variables and the
pvalue associated with the null hypothesis test (that the estimated correlation assu
mes value zero and is, therefore, not signicant). The rst column presents the
correlation results, considering the variable log (QTobin) with all other variables.
The results indicate that the log (market value) of the rms have a negative linear
association with the variables: log(R&D), in, log(capex), log(ATV), log(LT_inv),
log(ST_inv) gdp_current, and gdp_exp. Correlation measures of such variables
showed signs of statistical significance (rejecting the null hypothesis at 1%).
Although some variables, such as investments in R&D and longterm, have reec
ted negative signs on the correlation measurements, these results are incomplete
to dene more appropriately the relation between variables, since the correlation
does not impute causeandeffect (only linear association).
Estudios económicos N° 76, Enero  Junio 2020. 147195168
ESTUDIOS ECONOMICOS
Table 4. Crossdistribution of rms by geographic location and industry
Sector  SIC codes
Geographic location
Africa /
Middle
East
Asia /
Pacic Caribbean
Central
America
and
Mexico
Europe
Latin
America
and
Caribbean
United
States and
Canada
Total
Division A: Agriculture, Forestry,
and Fishing 10 122 1 0 32 733 205
Division B: Mining 74 901 31 6357 26 1 776 3 171
Division C: Construction 35 408 03 148 10 75 679
Division D: Manufacturing 535 713 32 26 2 034 146 2 806 12 709
Division E: Transportation,
Communications, Electric, Gas,
and Sanitary Services
117 853 36 19 520 83 505 2 133
Division F: Wholesale Trade 59 610 30 170 8 233 1 083
Division G: Retail Trade 58 502 5 8 235 22 344 1 174
Division H: Finance, Insurance,
and Real Estate 579 2 208 70 25 1 424 113 4 067 8 486
Division I: Services 136 1 615 17 2 102 24 1 363 4 177
Division J: Public Administration 42 145 6 2 36 9137 377
Total 1 645 14 494 201 91 5 976 448 11 339 34 194
Source: Authors’ calculations
Estudios económicos N° 76, Enero  Junio 2020. 147195 169
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Table 4. Crossdistribution of rms by geographic location and industry
Sector  SIC codes
Geographic location
Africa /
Middle
East
Asia /
Pacic Caribbean
Central
America
and
Mexico
Europe
Latin
America
and
Caribbean
United
States and
Canada
Total
Division A: Agriculture, Forestry,
and Fishing 10 122 1 0 32 733 205
Division B: Mining 74 901 31 6357 26 1 776 3 171
Division C: Construction 35 408 03 148 10 75 679
Division D: Manufacturing 535 713 32 26 2 034 146 2 806 12 709
Division E: Transportation,
Communications, Electric, Gas,
and Sanitary Services
117 853 36 19 520 83 505 2 133
Division F: Wholesale Trade 59 610 30 170 8 233 1 083
Division G: Retail Trade 58 502 5 8 235 22 344 1 174
Division H: Finance, Insurance,
and Real Estate 579 2 208 70 25 1 424 113 4 067 8 486
Division I: Services 136 1 615 17 2 102 24 1 363 4 177
Division J: Public Administration 42 145 6 2 36 9137 377
Total 1 645 14 494 201 91 5 976 448 11 339 34 194
Source: Authors’ calculations
Investments in research showed an inverse linear association, in addition
to the log (QTobin), with ination, the growth rate, and the balance of public
budget/GDP. The calculated measures presented statistical signicance at 1%. It is
necessary to note that the extent of correlation between log (QTobin) and ination
showed a low magnitude. However, log(R&D) and ination presented a much
higher magnitude, equivalent to four times compared to the previous one. It deno
tes a greater sensitivity of association of the investment in relation to the market
value of rms. Table 5 shows the correlation matrix between the model variables
and their respective pvalues.
IV.2. Results of the Econometric Model
Table 6 presents the results of model ME.1, excluding the quadratic varia
ble of ination. The isolated effect of investments in R&D was positive in most
columns, except for column (2). The isolated elasticity showed results between
0.05% and 0.18%  signicant at 1%. With the inclusion of nancial variables,
the coefcients demonstrated a closer and more stable values with little variation
(between 0.114% and 0.12%). Between the columns (2) to (6), inclusion and exclu
sion of nancial variables presented higher reections in the range of possibilities
of the coefcients, demonstrating a sensitivity to the microdimension variables
of rms.
Except for columns (2) and (6), ination had a positive effect on the market
value of rms (signicant parameters to 1%, except in columns (5) and (6)). The
inclusion/exclusion of microdimension variables showed an effect of underestima
ting the impact of ination, signaling for estimates below the parameter obtained
in the complete model (column (1)). In the opposite direction, the inclusion/exclu
sion of variables of macro dimension demonstrated an effect of overestimating the
impact of ination in comparison with the complete model.
Regarding the crosseffect between ination and investments in R&D, the
complete model showed a parameter with a negative and signicant sign at 1%.
Columns (2), (3), and (6) had a positive effect, however not signicant in column
(3). The other columns demonstrated negative parameters, but not signicant for
columns (4) and (5). These results indicate a volatility in the parameters, as we
include/exclude the microdimension variables, thus revealing a sensitivity with
such variables. With the inclusion/exclusion of macrodimension variables, the
coefcient obtained revealed signs of underestimation in relation to the complete
model (values in module in columns (8), (9), and (10)).
Estudios económicos N° 76, Enero  Junio 2020. 147195170
ESTUDIOS ECONOMICOS
Table 5. correlation matrix between the model variables and their respective pvalues
log
(QTobin)
log
(R&D) in log
(capex)
log
(ATV)
log (LT_
inv)
log (ST_
inv)
gdp_
growth
gdp_
budget
gdp_
current gdp_exp
log
(QTobin)
1.0000

log (R&D)
0.0343 1.0000
0.0000 
in
0.0745 0.3123 1.0000
0.0000 0.0000 
log (capex)
0.1360 0.5503 0.0019 1.0000
0.0000 0.0000 0.5005 
log (ATV)
0.4334 0.6423 0.0107 0.7910 1.0000
0.0000 0.0000 0.0000 0.0000 
log (LT_
inv)
0.2669 0.4912 0.0945 0.4567 0.7297 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 
log (ST_
inv) 0.2352 0.6869 0.1630 0.6741 0.8504 0.5746 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_growth 0.0872 0.1583 0.4998 0.0537 0.0238 0.0935 0.0181 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_budget 0.0092 0.0352 0.0814 0.0323 0.0363 0.0546 0.0732 0.1921 1.0000
0.0008 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_current 0.0671 0.0350 0.1609 0.0308 0.1022 0.0228 0.1667 0.1761 0.5995 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_exp 0.0531 0.0938 0.2254 0.0743 0.1345 0.0062 0.2157 0.1863 0.4753 0.8555 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0637 0.0000 0.0000 0.0000 0.0000 
Source: Authors’ calculations.
Note: Values in bold italics represent the pvalues associated with hypothesis testing.
Estudios económicos N° 76, Enero  Junio 2020. 147195 171
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Table 5. correlation matrix between the model variables and their respective pvalues
log
(QTobin)
log
(R&D) in log
(capex)
log
(ATV)
log (LT_
inv)
log (ST_
inv)
gdp_
growth
gdp_
budget
gdp_
current gdp_exp
log
(QTobin)
1.0000

log (R&D) 0.0343 1.0000
0.0000 
in 0.0745 0.3123 1.0000
0.0000 0.0000 
log (capex) 0.1360 0.5503 0.0019 1.0000
0.0000 0.0000 0.5005 
log (ATV) 0.4334 0.6423 0.0107 0.7910 1.0000
0.0000 0.0000 0.0000 0.0000 
log (LT_
inv) 0.2669 0.4912 0.0945 0.4567 0.7297 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 
log (ST_
inv)
0.2352 0.6869 0.1630 0.6741 0.8504 0.5746 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_growth
0.0872 0.1583 0.4998 0.0537 0.0238 0.0935 0.0181 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_budget
0.0092 0.0352 0.0814 0.0323 0.0363 0.0546 0.0732 0.1921 1.0000
0.0008 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_current
0.0671 0.0350 0.1609 0.0308 0.1022 0.0228 0.1667 0.1761 0.5995 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 
gdp_exp
0.0531 0.0938 0.2254 0.0743 0.1345 0.0062 0.2157 0.1863 0.4753 0.8555 1.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0637 0.0000 0.0000 0.0000 0.0000 
Source: Authors’ calculations.
Note: Values
in bold italics represent the pvalues
associated with hypothesis testing.
Estudios económicos N° 76, Enero  Junio 2020. 147195172
ESTUDIOS ECONOMICOS
Table 6. Model results. Dependent variable: log (QTobin)
Var. independent
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
log (R&D)
0.118*** 0.0574*** 0.0947*** 0.188*** 0.0628*** 0.0540*** 0.120*** 0.117*** 0.114*** 0.116***
(0.00812) (0.00635) (0.00688) (0.00704) (0.00726) (0.00732) (0.00806) (0.00847) (0.00849) (0.00849)
in
0.0438*** 0.0274*** 0.0267*** 0.0298*** 0.00614 0.0103 0.0602*** 0.0890*** 0.0634*** 0.0804***
(0.00850) (0.00751) (0.00783) (0.00711) (0.00860) (0.00765) (0.00789) (0.00865) (0.00885) (0.00937)
log (R&D) * in
0.00815*** 0.0113*** 0.00105 0.00125 0.000114 0.00768*** 0.00845*** 0.00599** 0.00492** 0.00525**
(0.00219) (0.00183) (0.00201) (0.00181) (0.00217) (0.00188) (0.00219) (0.00243) (0.00241) (0.00243)
Var. nancial
(microlevel)
            
                     
               
                        
log (capex)
0.0868*** 0.143*** 0.0883*** 0.0983*** 0.0990*** 0.100***
(0.00786) (0.00420) (0.00791) (0.00815) (0.00808) (0.00813)
log (ATV)
0.498*** 0.299*** 0.494*** 0.500*** 0.508*** 0.504***
(0.0159) (0.00660) (0.0160) (0.0164) (0.0163) (0.0164)
log (LT_inv)
0.0116** 0.0641*** 0.0114** 0.00852* 0.00924* 0.00887*
(0.00487) (0.00438) (0.00488) (0.00508) (0.00503) (0.00505)
log (ST_inv)
0.237*** 0.134*** 0.234*** 0.238*** 0.241*** 0.239***
(0.0109) (0.00702) (0.0109) (0.0111) (0.0110) (0.0111)
Var.
macroeconomic
(macrolevel)
           
                                                          
gdp_growth 0.0884*** 0.0913*** 0.100*** 0.0975*** 0.0820*** 0.0959*** 0.103***
(0.00573) (0.00567) (0.00552) (0.00512) (0.00588) (0.00555) (0.00529)
gdp_budget 0.00725** 0.0145*** 0.00909*** 0.00776** 0.00882** 0.0115*** 0.00848***
(0.00362) (0.00360) (0.00335) (0.00309) (0.00370) (0.00346) (0.00321)
gdp_current 0.0334*** 0.0307*** 0.0353*** 0.0347*** 0.0320*** 0.0323*** 0.0264***
(0.00423) (0.00401) (0.00380) (0.00364) (0.00450) (0.00387) (0.00302)
gdp_exp 0.00635*** 0.000684 0.00748*** 0.00720*** 0.00857*** 0.00416** 0.00430**
(0.00229) (0.00216) (0.00200) (0.00185) (0.00238) (0.00208) (0.00177)
Constant 1.248*** 0.371*** 0.166 1.200*** 0.445*** 0.00113 1.096*** 1.489*** 1.569*** 1.504***
(0.121) (0.116) (0.122) (0.126) (0.116) (0.121) (0.115) (0.121) (0.125) (0.125)
R20.332 0.185 0.236 0.359 0.241 0.221 0.323 0.296 0.306 0.296
R2 Adj 0.331 0.184 0.236 0.359 0.240 0.221 0.323 0.295 0.305 0.296
Sample 205 164 205 164 205 164 205 164 205 164 205 164 205 164 205 164 205 164 205 164
F test 127.2*** 155.8*** 151.5*** 222.7*** 124.0*** 156.9*** 133.4*** 121.6*** 125.0*** 122.4***
Fixed Effects
(sector, region, and
year)
Yes Ye s Ye s Ye s Ye s Ye s Ye s Ye s Ye s Yes
F Fixed Effects Test 53.79*** 98.69*** 53.92*** 60.12*** 65.71*** 76.92*** 74.20*** 71.16*** 75.29*** 73.84***
Source: Authors’ calculations.
Legend: *** p <0.01, ** p <0.05, * p <0.1. The parameters of the standard error estimates are robust for the presence of heteroscedasticity and
autocorrelation.
Estudios económicos N° 76, Enero  Junio 2020. 147195 173
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Table 6. Model results. Dependent variable: log (QTobin)
Var. independent (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
log (R&D) 0.118*** 0.0574*** 0.0947*** 0.188*** 0.0628*** 0.0540*** 0.120*** 0.117*** 0.114*** 0.116***
(0.00812) (0.00635) (0.00688) (0.00704) (0.00726) (0.00732) (0.00806) (0.00847) (0.00849) (0.00849)
in 0.0438*** 0.0274*** 0.0267*** 0.0298*** 0.00614 0.0103 0.0602*** 0.0890*** 0.0634*** 0.0804***
(0.00850) (0.00751) (0.00783) (0.00711) (0.00860) (0.00765) (0.00789) (0.00865) (0.00885) (0.00937)
log (R&D) * in 0.00815*** 0.0113*** 0.00105 0.00125 0.000114 0.00768*** 0.00845*** 0.00599** 0.00492** 0.00525**
(0.00219) (0.00183) (0.00201) (0.00181) (0.00217) (0.00188) (0.00219) (0.00243) (0.00241) (0.00243)
Var. nancial
(microlevel)             
                     
               
                        
log (capex) 0.0868*** 0.143*** 0.0883*** 0.0983*** 0.0990*** 0.100***
(0.00786) (0.00420) (0.00791) (0.00815) (0.00808) (0.00813)
log (ATV) 0.498*** 0.299*** 0.494*** 0.500*** 0.508*** 0.504***
(0.0159) (0.00660) (0.0160) (0.0164) (0.0163) (0.0164)
log (LT_inv) 0.0116** 0.0641*** 0.0114** 0.00852* 0.00924* 0.00887*
(0.00487) (0.00438) (0.00488) (0.00508) (0.00503) (0.00505)
log (ST_inv) 0.237*** 0.134*** 0.234*** 0.238*** 0.241*** 0.239***
(0.0109) (0.00702) (0.0109) (0.0111) (0.0110) (0.0111)
Var.
macroeconomic
(macrolevel)
           
                                                          
gdp_growth
0.0884*** 0.0913*** 0.100*** 0.0975*** 0.0820*** 0.0959*** 0.103***
(0.00573) (0.00567) (0.00552) (0.00512) (0.00588) (0.00555) (0.00529)
gdp_budget
0.00725** 0.0145*** 0.00909*** 0.00776** 0.00882** 0.0115*** 0.00848***
(0.00362) (0.00360) (0.00335) (0.00309) (0.00370) (0.00346) (0.00321)
gdp_current
0.0334*** 0.0307*** 0.0353*** 0.0347*** 0.0320*** 0.0323*** 0.0264***
(0.00423) (0.00401) (0.00380) (0.00364) (0.00450) (0.00387) (0.00302)
gdp_exp
0.00635*** 0.000684 0.00748*** 0.00720*** 0.00857*** 0.00416** 0.00430**
(0.00229) (0.00216) (0.00200) (0.00185) (0.00238) (0.00208) (0.00177)
Constant
1.248*** 0.371*** 0.166 1.200*** 0.445*** 0.00113 1.096*** 1.489*** 1.569*** 1.504***
(0.121) (0.116) (0.122) (0.126) (0.116) (0.121) (0.115) (0.121) (0.125) (0.125)
R20.332 0.185 0.236 0.359 0.241 0.221 0.323 0.296 0.306 0.296
R2 Adj 0.331 0.184 0.236 0.359 0.240 0.221 0.323 0.295 0.305 0.296
Sample
205 164 205 164 205 164 205 164 205 164 205 164 205 164 205 164 205 164 205 164
F test
127.2*** 155.8*** 151.5*** 222.7*** 124.0*** 156.9*** 133.4*** 121.6*** 125.0*** 122.4***
Fixed Effects
(sector, region, and
year)
Yes Ye s Ye s Ye s Ye s Ye s Ye s Ye s Ye s Yes
F Fixed Effects Test
53.79*** 98.69*** 53.92*** 60.12*** 65.71*** 76.92*** 74.20*** 71.16*** 75.29*** 73.84***
Source: Authors’ calculations.
Legend: *** p <0.01, ** p <0.05, * p <0.1. The parameters of the standard error estimates are robust for the presence of heteroscedasticity and
autocorrelation.
Estudios económicos N° 76, Enero  Junio 2020. 147195174
ESTUDIOS ECONOMICOS
Such results attest to the importance of ination in the rms’ value strategy,
negatively affecting the elasticity coefcient of investments in research. For a cou
ntry with a high ination rate of more than 10% per year, this implies reducing the
elasticity coefcient in R&D to values below 0.04%. Countries with low ination,
like an amount close to 2% per year, allow an elasticity coefcient around 0.10%.
However, the nal effect of ination may show nonlinear signals that are not pre
cisely captured in the present model. In this case, the quadratic share of ination
is shown in Table 7.
Considering the complete model, the elasticity coefcient of investments
in capital goods presented a value below the isolated parameter of investments in
R&D (0.09% against 0.12%). This parameter demonstrated relatively close values
along the columns (7)  (10), when dimension variables at macro level are inclu
ded/excluded (signicant parameters at 1%). The most discrepant value was that
in column (3), as we exclude variables of companies’ nancial dimension. This
demonstrates a relative sensitivity of the variables in the micro context of the rms.
Observing the short and longterm investments, the elasticity coefcient of
the variable ‘log (ST_inv)’ was higher in relation to the ‘log (LT_inv)’. This aspect
was also observed throughout the other columns, with the exception of column (5)
and (6), which showed negative signs to the variables, although signicant at 1%.
This fact can be attributed to the exclusion of the other nancial variables that have
a signicant inuence on the parameter, attesting to a bias in the parameters with
their eliminations.
In relation to macroeconomic variables, the growth rate of the economy
showed a positive relation in the market value of the rms across all columns
(signicant at 1%). The parameter obtained in columns (2)  (6), as it excludes/
includes variables at the nancial level, has an overestimating effect in relation to
the complete model in column (1)  except for column (5).
The public budget demonstrated a positive relationship in the market value
of rms, in all columns with signs of statistical signicance at levels between 1%
and 5%. The exclusion/inclusion of variables at the nancial level demonstrated
an overestimated effect on the parameter. This effect was also observed in column
(8), when the same procedure is applied to variables at the macro level. The current
account balance in relation to GDP had a negative impact on the market value of
rms (signicant at 1% in all columns). In the opposite direction, the current account
balance in relation to exports had a positive impact on the market value of rms
(signicant between 1% and 5%, except for column (2)). The model’s explanatory
Estudios económicos N° 76, Enero  Junio 2020. 147195 175
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
capacity showed values close to the adjusted one, signaling qualitative aspects to the
model and the variables used (signicant F test). The xed effects demonstrated a
signicant inuence in all columns (joint signicance at the level of 1%).
Table 7 presents the results of the ME.1 model, adding the effect of the qua
dratic variable of ination. As can be observed, the isolated effect of investments
in R&D was positive in most columns, again except for column (2)  signicant
parameters at 1% in all columns. The isolated elasticities revealed results between
0.05% and 0.19%. With the inclusion of the nancial variables (columns (7) 
(10)), the coefcients indicated closer and more stable values with little variation
(between 0.122% and 0.126%  variation of 0.006 p.p.).
Between columns (2) to (6), the inclusion and exclusion of nancial varia
bles showed greater reexes in the differences between the coefcients, indicating
a greater sensitivity of investments in R&D with the microdimension variables of
the rms. In relation to the crosseffect R&D/ination, the estimated parameters
presented negative values (except for columns (2) and (6), as we vary the inclusion/
exclusion of nancial variables). Except for columns (3) and (5), the estimated para
meters were statistically signicant. Considering the complete model, an economy
with high ination of approximately 10% implies an elasticity coefcient in R&D
close to 0.03%. Countries with ination below 2% administer an elasticity coefcient
above 0.10%, placing a value equal to the previous model. The coefcient associated
with the cross effect R&D/ination showed subtle differences between columns (7)
to (10), as the inclusion/exclusion of variables in the macro dimension varies.
An economy with high ination of approximately 10% implies an elasticity
coefcient in R&D close to 0.03%. Countries with a low of 2% ination show an
elastic coefcient above 0.10%, reaching a value equal to the previous model. The
coefcient associated with R&D effects crossination showed subtle differences
between the columns (7) to (10), in that varying the inclusion/exclusion of variables
in the macro scale.
In most columns, ination (linear component) had a positive effect on the
market value of rms (signicant parameters at 1%, except for column (2)). On the
other hand, the quadratic component had a negative and signicant effect at 1% in
all columns. The two effects taken together, signal for a nonlinear relationship of
ination and with a concavity downwards, indicating a maximum point in the esti
mated function. This quadratic form indicates that low levels of ination can serve
as an incentive in the value strategy of rms, with an increasing pattern between
variables. In turn, higher levels (beyond the maximum point) can negatively affect
Estudios económicos N° 76, Enero  Junio 2020. 147195176
ESTUDIOS ECONOMICOS
Table 7. Model results. Dependent variable: log (QTobin)
Var. independent
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
log (R&D)
0.123*** 0.0514*** 0.100*** 0.193*** 0.0669*** 0.0602*** 0.126*** 0.126*** 0.122*** 0.125***
(0.00823) (0.00679) (0.00715) (0.00733) (0.00745) (0.00763) (0.00819) (0.00854) (0.00861) (0.00858)
in
0.0969*** 0.0369** 0.0889*** 0.0921*** 0.0574*** 0.0553*** 0.116*** 0.180*** 0.147*** 0.174***
(0.0150) (0.0178) (0.0162) (0.0151) (0.0164) (0.0172) (0.0149) (0.0175) (0.0178) (0.0181)
log (R&D) * in
0.00957*** 0.00892*** 0.00114 0.00359* 0.00168 0.00520** 0.0101*** 0.00893*** 0.00785*** 0.00849***
(0.00226) (0.00209) (0.00217) (0.00203) (0.00227) (0.00212) (0.00226) (0.00245) (0.00246) (0.00247)
in
20.00535*** 0.00670*** 0.00644*** 0.00649*** 0.00524*** 0.00684*** 0.00598*** 0.0101*** 0.00910*** 0.0101***
(0.00160) (0.00203) (0.00181) (0.00173) (0.00186) (0.00197) (0.00168) (0.00209) (0.00211) (0.00214)
Var. nancial (micro
level)                                                                                
log (capex)
0.0866*** 0.143*** 0.0880*** 0.0971*** 0.0974*** 0.0984***
(0.00784) (0.00420) (0.00788) (0.00807) (0.00801) (0.00805)
log (ATV)
0.499*** 0.299*** 0.495*** 0.502*** 0.509*** 0.504***
(0.0159) (0.00660) (0.0160) (0.0163) (0.0162) (0.0163)
log (LT_inv)
0.0115** 0.0643*** 0.0117** 0.00927* 0.0100** 0.00954*
(0.00486) (0.00438) (0.00487) (0.00503) (0.00500) (0.00500)
log (ST_inv)
0.237*** 0.134*** 0.235*** 0.239*** 0.242*** 0.240***
(0.0109) (0.00702) (0.0109) (0.0111) (0.0110) (0.0110)
Var. macroeconomic
(macrolevel)                                                                                
gdp_growth 0.0834*** 0.0862*** 0.0949*** 0.0925*** 0.0775*** 0.0906*** 0.0965***
(0.00567) (0.00559) (0.00545) (0.00505) (0.00581) (0.00548) (0.00527)
gdp_budget 0.00479 0.0128*** 0.00725** 0.00612** 0.00668* 0.00972*** 0.00479
(0.00364) (0.00361) (0.00336) (0.00309) (0.00371) (0.00347) (0.00322)
gdp_current 0.0329*** 0.0299*** 0.0346*** 0.0339*** 0.0314*** 0.0316*** 0.0243***
(0.00424) (0.00403) (0.00381) (0.00365) (0.00451) (0.00388) (0.00299)
gdp_exp 0.00754*** 0.00170 0.00854*** 0.00819*** 0.00956*** 0.00519** 0.00281
(0.00231) (0.00217) (0.00201) (0.00185) (0.00240) (0.00209) (0.00176)
Constant 1.186*** 0.455*** 0.247** 1.118*** 0.508*** 0.0873 1.059*** 1.371*** 1.467*** 1.385***
(0.121) (0.117) (0.122) (0.126) (0.116) (0.122) (0.115) (0.121) (0.124) (0.123)
R20.333 0.186 0.238 0.360 0.242 0.223 0.326 0.303 0.311 0.303
R2 Adj 0.332 0.186 0.237 0.360 0.241 0.222 0.325 0.302 0.310 0.302
Sample 205,164 205,164 205,164 205,164 205,164 205,164 205,164 205,164 205,164 205,164
F test 124.0*** 150.0*** 147.0*** 215.4*** 119.9*** 151.3*** 130.8*** 122.8*** 125.3*** 123.0***
Fixed Effects (sector,
region, and year) Yes Ye s Ye s Yes Yes Ye s Ye s Ye s Ye s Yes
F Fixed Effects Test 57.19*** 103.54*** 58.41*** 66.19*** 68.77*** 82.40*** 75.41*** 77.90*** 80.86*** 80.51***
Source: Authors’ calculations.
Legend: *** p <0.01, ** p <0.05, * p <0.1. The parameters of the standard error estimates are robust for the presence of heteroscedasticity and
autocorrelation.
Estudios económicos N° 76, Enero  Junio 2020. 147195 177
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Table 7. Model results. Dependent variable: log (QTobin)
Var. independent (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
log (R&D) 0.123*** 0.0514*** 0.100*** 0.193*** 0.0669*** 0.0602*** 0.126*** 0.126*** 0.122*** 0.125***
(0.00823) (0.00679) (0.00715) (0.00733) (0.00745) (0.00763) (0.00819) (0.00854) (0.00861) (0.00858)
in 0.0969*** 0.0369** 0.0889*** 0.0921*** 0.0574*** 0.0553*** 0.116*** 0.180*** 0.147*** 0.174***
(0.0150) (0.0178) (0.0162) (0.0151) (0.0164) (0.0172) (0.0149) (0.0175) (0.0178) (0.0181)
log (R&D) * in 0.00957*** 0.00892*** 0.00114 0.00359* 0.00168 0.00520** 0.0101*** 0.00893*** 0.00785*** 0.00849***
(0.00226) (0.00209) (0.00217) (0.00203) (0.00227) (0.00212) (0.00226) (0.00245) (0.00246) (0.00247)
in20.00535*** 0.00670*** 0.00644*** 0.00649*** 0.00524*** 0.00684*** 0.00598*** 0.0101*** 0.00910*** 0.0101***
(0.00160) (0.00203) (0.00181) (0.00173) (0.00186) (0.00197) (0.00168) (0.00209) (0.00211) (0.00214)
Var. nancial (micro
level)                                                                                
log (capex) 0.0866*** 0.143*** 0.0880*** 0.0971*** 0.0974*** 0.0984***
(0.00784) (0.00420) (0.00788) (0.00807) (0.00801) (0.00805)
log (ATV) 0.499*** 0.299*** 0.495*** 0.502*** 0.509*** 0.504***
(0.0159) (0.00660) (0.0160) (0.0163) (0.0162) (0.0163)
log (LT_inv) 0.0115** 0.0643*** 0.0117** 0.00927* 0.0100** 0.00954*
(0.00486) (0.00438) (0.00487) (0.00503) (0.00500) (0.00500)
log (ST_inv) 0.237*** 0.134*** 0.235*** 0.239*** 0.242*** 0.240***
(0.0109) (0.00702) (0.0109) (0.0111) (0.0110) (0.0110)
Var. macroeconomic
(macrolevel)
                                                                               
gdp_growth
0.0834*** 0.0862*** 0.0949*** 0.0925*** 0.0775*** 0.0906*** 0.0965***
(0.00567) (0.00559) (0.00545) (0.00505) (0.00581) (0.00548) (0.00527)
gdp_budget
0.00479 0.0128*** 0.00725** 0.00612** 0.00668* 0.00972*** 0.00479
(0.00364) (0.00361) (0.00336) (0.00309) (0.00371) (0.00347) (0.00322)
gdp_current
0.0329*** 0.0299*** 0.0346*** 0.0339*** 0.0314*** 0.0316*** 0.0243***
(0.00424) (0.00403) (0.00381) (0.00365) (0.00451) (0.00388) (0.00299)
gdp_exp
0.00754*** 0.00170 0.00854*** 0.00819*** 0.00956*** 0.00519** 0.00281
(0.00231) (0.00217) (0.00201) (0.00185) (0.00240) (0.00209) (0.00176)
Constant
1.186*** 0.455*** 0.247** 1.118*** 0.508*** 0.0873 1.059*** 1.371*** 1.467*** 1.385***
(0.121) (0.117) (0.122) (0.126) (0.116) (0.122) (0.115) (0.121) (0.124) (0.123)
R
20.333 0.186 0.238 0.360 0.242 0.223 0.326 0.303 0.311 0.303
R
2
Adj
0.332 0.186 0.237 0.360 0.241 0.222 0.325 0.302 0.310 0.302
Sample
205,164 205,164 205,164 205,164 205,164 205,164 205,164 205,164 205,164 205,164
F test
124.0*** 150.0*** 147.0*** 215.4*** 119.9*** 151.3*** 130.8*** 122.8*** 125.3*** 123.0***
Fixed Effects (sector,
region, and year)
Yes Ye s Ye s Ye s Ye s Ye s Ye s Ye s Ye s Yes
F Fixed Effects Test
57.19*** 103.54*** 58.41*** 66.19*** 68.77*** 82.40*** 75.41*** 77.90*** 80.86*** 80.51***
Source: Authors’ calculations.
Legend: *** p <0.01, ** p <0.05, * p <0.1. The parameters of the standard error estimates are robust for the presence of heteroscedasticity and
autocorrelation.
Estudios económicos N° 76, Enero  Junio 2020. 147195178
ESTUDIOS ECONOMICOS
the forecast of the rm’s assets leading to a decreasing pattern with the market
value. The inclusion/exclusion of microdimension variables showed an effect of
underestimating the impact of ination, signaling for estimates below the parame
ter obtained in the complete model. On the other hand, the inclusion/exclusion of
variables of macro dimension demonstrated an effect of overestimating the impact
of ination in comparison to the same model.
For the other variables in the nancial dimension, the parameter associated
with investments in capital goods presented an elasticity coefcient of approximately
0.087% in the complete model (signicant at 1%). A negative value (column (3)) was
observed as we varied the inclusion/exclusion of nancial variables, demonstrating
investment sensitivity in relation to the other characteristics of the rms. The varia
bles at the macro level did not show any major inuence on the differences between
the parameters, although indicating signs of signicance at 1% in all columns.
The rms’ size, captured by the log (assets), demonstrated an inverse rela
tion in the log (QTobin) in all the estimated models (signicant at 1%). Column
(4) is highlighted, since its result presented an estimate much lower than the other
columns. Again, short and longterm investments showed positive signs, with
statistical signicance in most columns (except for only columns (5) and (6), which
presented negative signs).
In relation to macroeconomic variables, the growth rate of the economy
revealed a positive relation in the market value of the rms across all columns
(signicant at 1%). The estimated parameters revealed subtle differences between
the columns, indicating little dispersion in relation to the complete model of column
(1)  except the result obtained in column (5). Thus, countries with higher growth
rates make a greater contribution to the market value of their rms.
Again, the participation of the budgetary result in the GDP showed a posi
tive relation with the market value of the rms (signicant parameters at 1%, except
for column (8)). A greater difference between the estimates is highlighted as we
include/exclude variables of nancial dimension (columns (2) to (6)), signaling
greater sensitivity of the variable in relation to the characteristics of the rms. The
quality of the adjustment proved to be satisfactory with R2adjusted values close
to R2 (highlight for column (4), which presented higher values in both statistics,
followed by the complete model in column (1)). The tests of global signicance
of the model proved to be signicant at 1% in all models. We also highlight the
signicance of the xed effects in the estimated models, given that they rejected
the null hypothesis of no inuence on the model at the level of 1%.
Estudios económicos N° 76, Enero  Junio 2020. 147195 179
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Figure 1 demonstrates the inverse relation between the research elasticity
coefcients in the market value of rms and the ination rate between economies.
Relatively expressive levels can lead to a reduction in the coefcient until the pre
sence of negative levels in this relation.
Figures 2 and 3 illustrate the nonlinear effect of ination and the cross
inuence of investment in research. The increase in ination leads to a reduction
not only in the elasticity coefcient, but also directly in the value of rms. This
nonlinear form signals the tradeoff of the ination target policy, since controlled
ination values can lead to a positive effect on the rms’ value strategy.
As shown in Figure 2, higher levels of investment in R&D positively shift
the quadratic curve of ination, so that low levels of ination may still be associated
with a larger stock of rms’ value. This aspect may be related to economies with
greater intensity in innovation, whose investments are relatively more signicant,
visàvis the greater mechanisms of appropriateness of rms, although in scenarios
of lesser technological opportunity (Dosi, Marengo, & Pasquali, 2006).
Figure 1. Relationship between the coefcient of elasticity and ination
10 20 30 40 50 60
inflation
0.4
0.3
0.2
0.1
0.1
dlog Qtobin
dlog (R&D)
0
0
0
5
5
5
10
15
20
40
60
inflation
log (Qtobin)
inflation * log (R&D)
10 %
50 %
90 %
20 10
2
1
1
2
3
4
20 30 40
inflation
10
Source: Authors’ calculations.
Estudios económicos N° 76, Enero  Junio 2020. 147195180
ESTUDIOS ECONOMICOS
Figure 2. Relationship between the market value and ination (crosseffect)
10 20 30 40 50 60
inflation
0.4
0.3
0.2
0.1
0.1
dlog Qtobin
dlog (R&D)
0
0
0
5
5
5
10
15
20
40
60
inflation
log (Qtobin)
inflation * log (R&D)
10 %
50 %
90 %
20 10
2
1
1
2
3
4
20 30 40
inflation
10
Source: Authors’ calculations.
Figure 3. Relationship between the market value and ination
10 20 30 40 50 60
inflation
0.4
0.3
0.2
0.1
0.1
dlog Qtobin
dlog (R&D)
0
0
0
5
5
5
10
15
20
40
60
inflation
log (Qtobin)
inflation * log (R&D)
10 %
50 %
90 %
20 10
2
1
1
2
3
4
20 30 40
inflation
10
Source: Developed by the authors.
Note: To isolate the effect of ination, the estimated model of column (1), Table 7, was adopted as
the log value (R&D) function according to percentiles 10%, 50%, and 10% of the distribution of this
variable.
Estudios económicos N° 76, Enero  Junio 2020. 147195 181
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
IV.3. Dynamic Model Results
Table 8 presents the results of the model, including the quadratic effect of
the ination component, capturing a nonlinear relationship between ination and
the market value of rms.
Controlling the lagged effect of the rms’ value (L (t1) .log (QTobin)),
column (1) demonstrates the complete model, whose results pointed to a conver
gence in terms of the expected sign for the crosseffect between ‘ination x R&D’
(signicant at 1%). Contrary to previous results, the direct effect of ination proved
to be negative on the value of rms (signicant at 1%). The parameter associated
with the quadratic effect of ination showed a higher magnitude, indicating an
underestimation from the previous technique. Regarding the microlevel controls,
the expected signs were convergent with the previous results, with differences in
terms of magnitudes (signicant parameters). This same pattern was also observed
for variables at the macro level, with higher magnitudes in most variables (except
for GDPcurrent).
In relation to columns (2) to (4), the exclusion of variables at the micro level
demonstrated an inuence on the results of the parameters, in terms of the expected
sign and signicance, at least in the controls at the micro and the macrolevel (the
latter in columns (3) and (4)).
Based on columns (5) to (7), the exclusion/inclusion of variables at the
macro level also demonstrated signicant inuences on the other variables in the
model. These inuences are seen in the loss of signicance in the micro and
macrolevel controls, in addition to a signicant increase in the ination parameter
(absence of statistical signicance).
The Hansen J tests demonstrated that the selected instruments are valid
and not correlated with the stochastic disturbance (except for columns (2) and (3),
which showed statistical signicance at the level of 10%).
Exogeneity tests for subsets of the instruments revealed that they did not
reject the null hypothesis of validity. Finally, the residual autocorrelation tests
demonstrated not to reject the null hypothesis of absence in almost all columns
(except for columns (4) and (7)).
Estudios económicos N° 76, Enero  Junio 2020. 147195182
ESTUDIOS ECONOMICOS
Table 8. Results of the GMM System
Independente var.
(1) (2) (3) (4) (5) (6) (7)
L(t1)
.log(QTobin)
0.967*** 0.711*** 0.677*** 0.937*** 0.957*** 0.995*** 0.944***
(0.0583) (0.0287) (0.0358) (0.121) (0.101) (0.111) (0.108)
log (R&D)
0.0841*** 0.293*** 0.538** 0.513 0.508** 0.419 0.556*
(0.0250) (0.0786) (0.216) (0.324) (0.255) (0.299) (0.329)
in
0.0262*** 0.100*** 0.155*** 0.158* 0.0892 0.135 0.173
(0.0101) (0.0263) (0.0332) (0.0913) (0.0810) (0.107) (0.115)
log (R&D) * in
0.811*** 0.0519 0.250** 1.143*** 1.049** 1.321** 1.339**
(0.223) (0.0343) (0.109) (0.404) (0.484) (0.569) (0.577)
in
20.0890*** 0.00222 0.0139* 0.0789** 0.0787** 0.0999** 0.0959**
(0.0221) (0.00492) (0.00813) (0.0308) (0.0369) (0.0426) (0.0422)
Var. nancial (microlevel)
      
log (capex)
0.0419** 0.0668 0.00479 0.0686 0.0231 0.0830 0.106
(0.0188) (0.0514) (0.0244) (0.128) (0.116) (0.135) (0.136)
log (ATV)
0.183*** 0.166* 0.316*** 0.340* 0.421* 0.490**
(0.0465) (0.0899) (0.102) (0.204) (0.231) (0.246)
log (LT_inv)
0.0210** 0.0421 0.0347 0.0503 0.0516
(0.0104) (0.0452) (0.0304) (0.0352) (0.0354)
log (ST_inv)
0.0583*** 0.0123 0.132 0.126
(0.0100) (0.202) (0.259) (0.256)
Var. macroeconomic
(macrolevel)
      
gdp_growth 0.118*** 0.0108** 0.0390*** 0.169 0.108 0.175 0.163
(0.0438) (0.00551) (0.0151) (0.111) (0.0983) (0.122) (0.120)
gdp_budget 0.0209*** 0.0298*** 0.00949 0.0307 0.0593* 0.0348
(0.00602) (0.00860) (0.00671) (0.0252) (0.0332) (0.0292)
gdp_current 0.0205*** 0.0339*** 0.00966 0.0450 0.0370***
(0.00609) (0.00765) (0.00791) (0.0326) (0.0134)
gdp_exp 0.0153** 0.00858* 0.00424 0.0431**
(0.00705) (0.00471) (0.00303) (0.0197)
Instruments 16 12 13 13 11 12 13
Hansen Jtest 4.986 4.608* 5.091* 0.896 1.674 0.399 0.744
pvalue 0.173 0.0998 0.0784 0.344 0.196 0.528 0.388
DiffinHansen test 0.34 0.60 0.25 0.90 1.67 0.40 0.74
pvalue 0.558 0.440 0.617 0.344 0.196 0.528 0.388
ArellanoBond test       
 AR(1)  z test 8.319*** 18.94*** 14.34*** 4.469*** 5.318*** 3.918*** 3.636***
pvalue 0.000 0.000 0.000 0.000 0.000 0.000 0.000277
 AR(2)  z test 0.38 0.596 0.404 2.196** 1.574 0.96 2.089**
pvalue 0.705 0.551 0.686 0.0281 0.115 0.337 0.0367
Source: Elaborated by the authors
Note: e instrument matrix was ‘collapsed’ using the technique proposed by Roodman (2009), in order to avoid the proliferation of instruments that
would compromise the model’s results. Regarding the reported statistics, Hansen Jtest assesses, in the null hypothesis, that the instruments are valid.
e AR (1) and AR (2) tests are for rst and second order autocorrelated disturbances in the rst dierences equations. DiinHansen test reports the
validity of the additional moment restrictions that are necessary for GMM system
Estudios económicos N° 76, Enero  Junio 2020. 147195 183
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Table 8. Results of the GMM System
Independente var. (1) (2) (3) (4) (5) (6) (7)
L(t1).log(QTobin) 0.967*** 0.711*** 0.677*** 0.937*** 0.957*** 0.995*** 0.944***
(0.0583) (0.0287) (0.0358) (0.121) (0.101) (0.111) (0.108)
log (R&D) 0.0841*** 0.293*** 0.538** 0.513 0.508** 0.419 0.556*
(0.0250) (0.0786) (0.216) (0.324) (0.255) (0.299) (0.329)
in 0.0262*** 0.100*** 0.155*** 0.158* 0.0892 0.135 0.173
(0.0101) (0.0263) (0.0332) (0.0913) (0.0810) (0.107) (0.115)
log (R&D) * in 0.811*** 0.0519 0.250** 1.143*** 1.049** 1.321** 1.339**
(0.223) (0.0343) (0.109) (0.404) (0.484) (0.569) (0.577)
in20.0890*** 0.00222 0.0139* 0.0789** 0.0787** 0.0999** 0.0959**
(0.0221) (0.00492) (0.00813) (0.0308) (0.0369) (0.0426) (0.0422)
Var. nancial (microlevel)       
log (capex) 0.0419** 0.0668 0.00479 0.0686 0.0231 0.0830 0.106
(0.0188) (0.0514) (0.0244) (0.128) (0.116) (0.135) (0.136)
log (ATV) 0.183*** 0.166* 0.316*** 0.340* 0.421* 0.490**
(0.0465) (0.0899) (0.102) (0.204) (0.231) (0.246)
log (LT_inv) 0.0210** 0.0421 0.0347 0.0503 0.0516
(0.0104) (0.0452) (0.0304) (0.0352) (0.0354)
log (ST_inv) 0.0583*** 0.0123 0.132 0.126
(0.0100) (0.202) (0.259) (0.256)
Var. macroeconomic
(macrolevel)       
gdp_growth
0.118*** 0.0108** 0.0390*** 0.169 0.108 0.175 0.163
(0.0438) (0.00551) (0.0151) (0.111) (0.0983) (0.122) (0.120)
gdp_budget
0.0209*** 0.0298*** 0.00949 0.0307 0.0593* 0.0348
(0.00602) (0.00860) (0.00671) (0.0252) (0.0332) (0.0292)
gdp_current
0.0205*** 0.0339*** 0.00966 0.0450 0.0370***
(0.00609) (0.00765) (0.00791) (0.0326) (0.0134)
gdp_exp
0.0153** 0.00858* 0.00424 0.0431**
(0.00705) (0.00471) (0.00303) (0.0197)
Instruments
16 12 13 13 11 12 13
Hansen Jtest
4.986 4.608* 5.091* 0.896 1.674 0.399 0.744
pvalue
0.173 0.0998 0.0784 0.344 0.196 0.528 0.388
DiffinHansen test
0.34 0.60 0.25 0.90 1.67 0.40 0.74
pvalue
0.558 0.440 0.617 0.344 0.196 0.528 0.388
ArellanoBond test
      
 AR(1)  z test
8.319*** 18.94*** 14.34*** 4.469*** 5.318*** 3.918*** 3.636***
pvalue
0.000 0.000 0.000 0.000 0.000 0.000 0.000277
 AR(2)  z test
0.38 0.596 0.404 2.196** 1.574 0.96 2.089**
pvalue
0.705 0.551 0.686 0.0281 0.115 0.337 0.0367
Source: Elaborated by the authors
Note: e instrument matrix was ‘collapsed’ using the technique proposed by Roodman (2009), in order to avoid the proliferation of instruments that
would compromise the model’s results. Regarding the reported statistics, Hansen Jtest assesses, in the null hypothesis, that the instruments are valid.
e AR (1) and AR (2) tests are for rst and second order autocorrelated disturbances in the rst dierences equations. DiinHansen test reports the
validity of the additional moment restrictions that are necessary for GMM system
Estudios económicos N° 76, Enero  Junio 2020. 147195184
ESTUDIOS ECONOMICOS
IV.4. Discussion With Recent Researches
The results presented expose theoretically and empirically the negative
impacts of ination on innovation efforts and, consequently, on the rate of tech
nological progress, since investments in R&D are the main input for innovation
(Hall, Lotti, & Mairesse, 2013).
Recent evidences, highlighting Chu and Lai (2013), show that an increase of
1% in ination contributes to reduce the R&D/GDP ratio by approximately 0.026%.
Other studies point to close relation between ination and demand for R&D at the
aggregate level, highlighting Wu and Zhang (1998) and Ho, Zeng, and Zhang (2007).
Another recent contribution is the study of Chu et al. (2015), who analyzed the
impact of ination on demand for investment in research when the economy presents
cash restrictions by rms (cashinadvance constraint). Evidence from the model
shows that investments in R&D are signicantly inuenced by the cash ow of rms.
This constraint is affected by the opportunity costs of money that are determined by
ination in each country. Economies with high ination levels manage a low level
of investment, as well as greater volatility in growth rates. One of the transmission
channels in this effect is perceived in the labor market, since ination induces a
reallocation effect between research activities for productive activities less intensive
in innovation. Empirical results showed a negative and statistically signicant asso
ciation, so that an increase of 1% in ination contributes to reduce the intensity of
investments in R&D/GDP by approximately 0.374%. In ‘eurozone’, the estimated
semielasticity corresponds to a value of 0.448% and 0.266% for the USA.
An important recent contribution on the topic is the research of Chu and Ji
(2016), who assessed the effects of monetary policy on economic growth, social
welfare, and the endogenous market structure in different economies. The study’s
ndings suggest that monetary policy has a transitory, though not permanent, effect
on the rate of economic growth. Specically, an increase in the nominal interest
rate, due to an increase in currency growth, reduces the level of output in the balan
ced growth path, but without a persistent effect on the steadystate growth rate.
The absence of currency neutrality with respect to the product level is asso
ciated with the market structure of the economy that responds endogenously to
changes in the labor market (specically labor supply) that are induced by monetary
policy. The market share that each rm holds is determined endogenously by the
conditions of entry and exit in response to the macroeconomic environment. Thus,
the propensity of rms to invest in R&D depends on their respective market share,
Estudios económicos N° 76, Enero  Junio 2020. 147195 185
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
which is determined by the structure, that is, by the composition of other rms in
the market and not by the aggregate market. In the short term, an increase in nomi
nal interest rates causes a reduction in labor supply, negatively impacting the size
of the rm, efforts in innovation, product, and consumption. In the long run, the
increase in the interest rate reduces the steadystate variables at a level, but with no
effect on the growth rate, since it is endogenously offset by the market structure.
Other prominent studies, such as Funk and Kromen (2010) and Chu and
Cozzi (2014), investigated the effects of monetary policy on growth and well
being conditions through endogenous growth models based on R&D. As shown,
the instability of the economy affected by ination can considerably inuence
the efciency in the use of investments in R&D. This uncertainty associated with
changes in relative prices and, consequently, in the costs of inputs, has a direct
impact on the realization and returns of investments in research. This can lead to a
“delay” in investment decisions that expect more favorable conditions (economic
stability) for the execution of innovation projects. This delay effect is associated
with the contributions of Manseld (1980), whose ination generates an increase
in uncertainties that contaminate the economy and limit efforts in innovation. Thus,
the persistence of ination generates costs that will have an impact on the techno
logical trajectory of the most diverse economies.
V. STUDY LIMITATIONS AND SUGGESTIONS FOR FUTURE RESEARCH
Although this study pointed to signicant effects between ination and the
result of investments in R&D in the rms’ value strategy (through the elasticity
coefcient of investments in R&D), important aspects of the model have not been
properly addressed.
Ination constraints, such as price shocks and their endogenous responses,
random and climatic factors, in addition to the differences in nancial develop
ment observed between economies can considerably inuence ination and its
persistence over time, according to recent researches (Canarella & Miller, 2017;
Bratsiotis, Madsen, & Martin, 2015; Stein, 2012; Bhattarai, Lee, & Park, 2014;
Gilchrist et al., 2017).
Although this study did not deepen the debate on such issues, the empirical
exercise allowed us to analyze how ination, as well as monetary policy guided
by the target regime, can affect rms’ R&D investment strategy. The empirical
advance, through effective controls over differences in nancial development or
Estudios económicos N° 76, Enero  Junio 2020. 147195186
ESTUDIOS ECONOMICOS
even in the orientation of monetary policy, allows to improve the understanding of
the possible costs that affect the relationship between investments and the market
value of rms. Although due control in the empirical model was not the highlight
of the study, the results presented converge with recent studies (Burdekin et al.,
2004; Gillman & Kejak, 2005; LópezVillavicencio & Mignon, 2011; He & Zou,
2016; Ramzi & Viem, 2016; Brunnermeier & Sannikov, 2016).
Furthermore, part of the results of the model, that point to an effect of ination
on the relationship ‘investments in R&D versus rms value’, has its understanding
of causeeffect still limited, since problems of endogeneity in the demand for invest
ments in R&D was not properly treated in the study. Although outside the central
focus of the theoretical model, research on the subject reports the importance of
treatment in econometric models (see Mishra (2007); Pires (2009)). In addition to
endogeneity, uctuations in proximity to the border affect the allocation of nancial
resources, especially in investments in R&D. This implies different results depen
ding on the technological position of each rm (see the contributions of Acemoglu,
Aghion, & Zilibotti, 2006; Wu, 2010; Amable, Demmou, & Ledezma, 2010; Hölzl &
Janger, 2014; Ding, Sun, & Jiang, 2016; Rocha et al., 2018; and Rocha et al., 2019).
It should be noted the limitations in the macrodimension of the variables
that can impact the model in new results. Recent and important research (e.g. Chu et
al., 2020) highlights the direct effect of nancial development on the pattern of local
innovation. Thus, as in the present study, inationary differences between economies
may be linked to different patterns in nancial development, directly impacting the
“ination versus rm value” relationship. This association with development has also
been duly documented in Aghion, Howitt, and MayerFoulkes (2005).
In this way, future research can be oriented towards a better treatment of
such conditions, understanding in a more appropriate way the relationship between
ination and the result of investments in innovation. In this sense, such advances
can contribute to a better design of the policy, reducing its negative effects and
stimulating, in a sustainable and balanced way, future efforts aimed at innovation
in different economies.
VI. FINAL CONSIDERATIONS
This research analyzed how ination can affect innovation efforts and, con
sequently, technological progress in different economies. By building an endoge
nous Schumpeterian growth model, rms face signicant cash constraints to nance
Estudios económicos N° 76, Enero  Junio 2020. 147195 187
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
innovation projects. Part of the resources come from own source, of appropriate
prots from the results of the innovation, while another part is funded through
nancial institutions.
The results of the theoretical model demonstrate that, in the presence of rising
ination above the level established by the monetary authority, inevitable adjustments
in interest rates increase the opportunity costs of investments in research. Thereby, a
lower propensity to invest is veried as the costs of nancing rise, reducing the results
of innovation and, consequently, the rate of technological progress.
Based on such theoretical results, an empirical regression model constructed
with a sample of 34 194 rms between 2010 and 2015 demonstrated that ination
presents different results among economies. High levels of ination reduce the
elasticity coefcient of investments in R&D, providing a lower return on rms’
value strategy. Countries with ination levels above 10% per year tend to have, on
average, an elasticity coefcient in R&D below 0.03%. In economies with greater
price stability, with ination rates below 2% per year, the elasticity coefcient is
greater than 0.10% for rms.
In addition to the results in the elasticity coefcient, ination presented a
nonlinear pattern in its effect in relation to the market value (log (Qtobin)) of the
rms. Countries with low and moderate levels have a positive relation with the value
of rms, signaling greater adherence to the rms’ value strategy. However, higher
levels of ination reect less predictability of assets, reducing rms’ incentives and
signaling a decreasing relationship with market value. The combination of the two
effects implies a quadratic relationship with the concavity downwards, implying a
maximum level of ination that limits the relationships between the variables.
The dynamic model results are converging with the theoretical model and
with previous results, except for the direct effect of ination on the value of com
panies, which demonstrated a signicant change in the expected sign (positive to
negative). The different techniques adopted indicate a greater robustness of the
results, suggesting a negative effect of ination on the elasticity coefcient of
investments in R&D. It indicates that higher ination tends to compromise a con
siderable part of the incentives for rms to innovate.
The results presented are convergent in relation to the recent contributions
of Chu and Lai (2013), Chu and Cozzi (2014), Chu, Cozzi, and Furukawa (2014),
Chu et al. (2015), Oikawa and Ueda (2015), Chu and Ji (2016) and Chu et al.
(2017), as well as pioneering studies like Manseld (1980). What is veried is that
Estudios económicos N° 76, Enero  Junio 2020. 147195188
ESTUDIOS ECONOMICOS
investments in R&D are negatively affected by the ination present in economies,
discouraging innovation activities and reducing the rate of technological progress.
In view of the contributions presented, further research on ination factors among
economies is necessary to understand more clearly the potential costs associated
with the monetary policy effort in price stability.
REFERENCES
Acemoglu, D. (2009). Introduction to Modern Economic Growth. Princeton, New
Jersey: Princeton University Press.
Acemoglu, D., Aghion, p., & Zilibotti, f. (2006). Distance to Frontier, Selection and
Economic Growth. Journal of the European Economic Association, 4(1), 3774.
Aghion, P., & Howitt, P. (2009). e Economics of Growth. Cambridge, Massachu
setts: e MIT Press.
Aghion, P., & Jaravel, X. (2015). Knowledge Spillovers, Innovation and Growth. e
Economic Journal, 125(583), 533545.
Aghion, P., & Marinescu, I. (2007). Cyclical Budgetary Policy and Economic
Growth: What Do We Learn from OECD Panel Data?. In D. Acemoglu,
K. Rogo, & M. Woodford, NBER Macroeconomics Annual. (Vol. 22, pp.
251278). Chicago: University of Chicago Press.
Aghion, P., Akcigit, U., & Howitt, P. (2013). What Do We Learn From Schumpete
rian Growth eory? In P. Aghion, & S. Durlauf, Handbook of Economic
Growth (Vol. 2A, pp. 515563). North Holland: Elsevier.
Aghion, P., Hémous, D., & Kharroubi, E. (2014). Cyclical scal policy, credit con
straints, and industry growth. Journal of Monetary Economics, 64(C), 4158.
Aghion, P., Howitt, P., & Prantl, S. (2015). Patent rights, product market reforms,
and innovation. Journal of Economic Growth, 20(3), 223262.
Aliyu, S. (2012). Does ination have an impact on stock returns and volatility? Evi
dence from Nigeria and Ghana. Applied Financial Economics, 22(6), 427435.
AlonsoBorrego, C., & Arellano, M. (1999). Symmetrically normalized instrumen
talvariable estimation using panel data. Journal of Business & Economic
Statistics, 17(1), 3649.
Amable, B., Demmou, L., & Ledezma, I. (2010). Product market regulation, innovation
and distance to frontier. Industrial and Corporate Change, 19(1), 117159.
Anand, R., Prasad, E., & Zhang, B. (2015). What measure of ination should a
developing country central bank target? Journal of Monetary Economics,
74(C), 102116.
Arellano, M. (2016). Modelling optimal instrumental variables for dynamic panel
data models. Research in Economics, 70(2), 238261.
Estudios económicos N° 76, Enero  Junio 2020. 147195 189
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Arellano, M., & Bond, S. (1991). Some tests of specication for panel data: Monte
Carlo evidence and an application to employment equations. e Review
of Economic Studies, 58(2), 277297.
Arellano, M., & Bover, O. (1995). Another look at the instrumental variable estima
tion of errorcomponents models. Journal of Econometrics, 68(1), 2951.
Ascari, G., & Sbordone, A. (2014). e Macroeconomics of Trend Ination. Journal
of Economic Literature, 52(3), 679739.
Baltagi, B. (2005). Econometric Analysis of Panel Data (3ª ed.). West Sussex: John
Wiley & Sons Ltd.
Barro, R., & SalaIMartin, X. (2004). Economic Growth (2ª ed.). Cambridge, Mas
sachusetts: e MIT Press.
Bashir, A.H. (2002). e welfare eects of ination and nancial innovation in a
model of economic growth: An Islamic perspective. Journal of Economic
Studies, 29(1), 2132.
Beal, A., & Graham, L. (2014). Foundations for Change: Rule of Law, Development,
and Democratization. Politics & Policy, 42(3), 311345 .
Beck, T., & Levine, R. (2004). Stock markets, banks, and growth: panel evidence.
Journal of Banking and Finance, 28(3), 423442.
Becker, R., Orborn, D., & Yildirim, D. (2012). A threshold cointegration analysis
of interest rate passthrough to UK mortgage rates. Economic Modelling,
29(6), 25042513.
Bester, C., & Hansen, C. (2016). Grouped eects estimators in xed eects models.
Journal of Econometrics, 190(1), 197208.
Bhamra, H., Kuehn, L., & Strebulaev, I. (2010). Long run risks, credit markets, and
nancial structure. American Economic Review: Papers and Proceedings,
100(2), 547551.
Bhattarai, S., Lee, J., & Park, W. (2014). Ination dynamics: e role of public debt
and policy regimes. Journal of Monetary Economics, 67(C), 93108.
Binder, M., Hsiao, C., & Pesaran, M. (2005). Estimation and inference in short panel
vector autoregressions with unit roots and cointegration. Econometric e
ory, 21(4), 795837.
Blanchard, O. (2016). e Phillips Curve: Back to the ‘60s? American Economic
Review, 106(5), 3134.
Blundell, R., & Bond, S. (1998). Initial conditions and moment restrictions in
dynamic panel data models. Journal of Econometrics, 87(1), 115–143.
Bogliacino, F., & Cordona, S. (2010). e determinants of R&D Investment: the role
of Cash ow and Capabilities. (IPTS Working Papers on Corporate R&D
and Innovation 10/2010). Recuperado de https://www.econstor.eu/bit
stream/10419/202126/1/jrcwp201010.pdf
Estudios económicos N° 76, Enero  Junio 2020. 147195190
ESTUDIOS ECONOMICOS
Bourke, J., & Roper, S. (2017). Innovation, quality management and learning: Short
term and longerterm eects. Research Policy, 46(8), 15051518.
Bratsiotis, G., Madsen, J., & Martin, C. (2015). Ination targeting and ination
persistence. Economic and Political Studies, 3(1), 317.
Brunnermeier, M., & Sannikov, Y. (2016). On the Optimal Ination Rate. American
Economic Review, 106(5), 484489.
Burdekin, R., Denzau, A., Keil, M., Sitthiyot, T., & Willett, T. (2004). When does
ination hurt economic growth? Dierent nonlinearities for dierent
economies. Journal of Macroeconomics, 26(3), 519–532.
Canarella, G., & Miller, S. (2017). Ination targeting and ination persistence: New
evidence from fractional integration and cointegration. Journal of Econom
ics and Business, 92(C), 4562.
Capistrán, C., & RamosFrancia, M. (2010). Does Ination Targeting Aect the
Dispersion of Ination Expectations? Journal of Money, Credit and Bank
ing, 42(1), 113134.
Carkovic, M., & Levine, R. (2005). Does foreign direct investment accelerate eco
nomic growth? In T. Moran, E. Graham, & M. Blomström (Eds.). Does
Foreign Direct Investment Promote Development? (pp. 195220). Wash
ington, DC: Institute for International Economics and Center for Global
Development.
Charron, N. (2011). Exploring the impact of foreign aid on corruption: has the
“AntiCorruption Movement” been eective? e Developing Economies,
49(1), 6688.
Chen, L., CollinDufresne, P., & Goldstein, R. (2009). On the relation between the
credit spread puzzle and the equity premium puzzle. Review of Financial
Studies, 22(9), 33673409.
Chu, A., & Cozzi, G. (2014). R&D and economic growth in a cashinadvance
economy. International Economic Review, 55(2), 507524. Recuperado de
https://doi.org/10.1111/iere.12059
Chu, A., & Ji, L. (2016). Monetary policy and endogenous market structure in a
Schumpeterian economy. Macroeconomic Dynamics, 20(5), 11271145.
Chu, A., & Lai, C.C. (2013). Money and the Welfare Cost of Ination in an R&D
Growth Model. Journal of Money, Credit and Banking, 45(1), 233249.
Chu, A., Cozzi, C., Lai, C.C., & Liao, C.H. (2015). Ination, R&D and growth
in an open economy. Journal of International Economics, 96(2), 360374.
Chu, A., Cozzi, G., Fan, H., Pan, S., & Zhang, M. (2020). Do Stronger Patents Stim
ulate or Stie Innovation? e Crucial Role of Financial Development.
Journal of Money, Credit and Banking, 52(5), 13051322.
Chu, A., Cozzi, G., Furukawa, Y., & Liao, C.H. (2017). Ination and economic
growth in a Schumpeterian Model with endogenous entry of heterogenous
Estudios económicos N° 76, Enero  Junio 2020. 147195 191
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
rms. Munich Personal RePEc Archive (MPRA Paper No. 77543). Munich,
Germany: University Library of Munich.
Coad, A. (2011). Appropriate business strategy for leaders and laggards. Industrial
and Corporate Change, 20(4), 10491079.
Cremers, M., Pareek, A., & Sautner, Z. (2017). ShortTerm Investors, LongTerm
Investments, and Firm Value. Consultado el 05 15, 2017, disponible en
https://ssrn.com/abstract=2720248
Dias, A. (2013). Market capitalization and ValueatRisk. Journal of Banking &
Finance, 37(12), 52485260.
Ding, S., Sun, P., & Jiang, W. (2016). e Eect of Import Competition on Firm
Productivity and Innovation: Does the Distance to Technology Frontier
Matter? Oxford Bulletin of Economics and Statistics, 78(2), 03059049.
Dosi, G., Marengo, L., & Pasquali, C. (2006). How much should society fuel the
greed of innovators?: On the relations between appropriability, opportu
nities and rates of innovation. Research Policy, 35(8), 11101121.
Dressler, S. (2016). A longrun, shortrun, and politicoeconomic analysis of the
welfare costs of ination. Journal of Macroeconomics, 47(Part B), 255269.
Ege, A., & Ege, A. (2017). Is there an innovation eect on current account? A case
for European countries. Applied Economics, online, 113.
Funk, P., & Kromen, B. (2010). Ination and innovationdriven growth. e B.E.
Journal of Macroeconomics, 10(1), 19351690.
Gehringer, A. (2013). Another Look at the Determinants of Current Account Imbal
ances in the European Union: An Empirical Assesment. (FIW Working
Paper No. 105), Vienna: Federal Ministry of Science, Research and Econ
omy.
Gilchrist, S., Schoenle, R., Sim, J., & Zakrajsek, E. (2017). Ination Dynamics during
the Financial Crisis. American Economic Review, 107(3), 785823.
Gillman, M., & Kejak, M. (2005). Contrasting models of the eect of ination on
growth. Journal of Economic Surveys, 19(1), 113136 .
Gomes, J., & Schmid, L. (2010). Levered returns. Journal of Finance, 65(2), 467494.
Goodfriend, M. (2007). How the World Achieved Consensus on Monetary Policy.
Journal of Economic Perspectives, 21, 4768.
Gourio, F. (2013). Credit risk and disaster risk. American Economic Journal: Mac
roeconomics, 5(3), 134.
Graham, J., Campbell, R., & Rajgopal, S. (2005). e economic implications of corpo
rate nancial reporting. Journal of Accounting and Economics, 4(13), 373.
Greene, W. (2012). Econometric Analysis (7ª ed.). Boston, MA: Prentice Hall.
Gropp, R., Sørensen, C., & Lichtenberger, J.D. (2007). e dynamics of bank spread
and nancial structure. (ECB, Working Papers Series No. 714). Frankfurt:
Europen Central Bank
Estudios económicos N° 76, Enero  Junio 2020. 147195192
ESTUDIOS ECONOMICOS
Gupta, K., Banerjee, R., & Onur, I. (2017). e eects of R&D and competition on
rm value: International evidence. International Review of Economics &
Finance, 51(C), 391404.
Hackbarth, D., Miao, J., & Morellec, E. (2006). Capital structure, credit risk, and mac
roeconomic conditions. Journal of Financial Economics, 82(3), 519550.
Hahn, J., & Kuersteiner, G. (2002). Asymptotically unbiased inference for a dynamic
panel model with xed eects when both N and T are large. Econometrica,
70(4), 16391657.
Hahn, J., & Newey, W. (2004). Jackknife and analytical bias reduction for nonlinear
panel models. Econometrica, 72(4), 12951319.
Hal, B., Jae, A., & Trajtenberg, M. (2005). Market Value and Patent Citations. e
RAND Journal of Economics, 36(1), 1638.
Hall, B. (2002). e Financing of Research and Development. Oxford Review of
Economic, 18(1), 3551.
Hall, B., Jae, A., & Trajtenberg, M. (2005). Market Value and Patent Citations. e
RAND Journal of Economics, 36(1), 1638.
Hall, B., Lotti, F., & Mairesse, J. (2013). Evidence on the impact of R&D and ICT
investments on innovation and productivity in Italian rms. Economics of
Innovation and New Technology, 22(3), 300328.
Hall, B., Mairesse, J., & Mohnen, P. (2010). Measuring the Returns to R&D. In B.
Hall, & N. Rosenberg, Handbook of the Economics of Innovation (Vol. 2,
pp. 10331082). Amsterdan: NorthHolland by Elsevier.
Hall, R., & Jones, C. (1999). Why do some countries produce so much more output
per worker than others? Quarterly Journal of Economics, 114(1), 83116.
Hall, S., & UrgA, G. (2000). New developments in the analysis of panel data sets. In
S. DAHIYA (Ed.), e Current State of Business Disciplines (pp. 53764).
Rohtak: Spellbound Publications PVT Ltd. (Business Economics).
He, Q., & Zou, H.F. (2016). Does ination cause growth in the reformera China?
eory and evidence. International Review of Economics & Finance, 45(C),
470484.
Ho, W., Zeng, J., & Zhang, J. (2007). Ination taxation and welfare with externalities
and leisure. Journal of Money, Credit and Banking, 39(1), 105131.
Hölzl, W., & Janger, J. (2014). Distance to the frontier and the perception of inno
vation barriers across European countries. Research Policy, 43(4), 707725.
Hosny, A. (2017). Does ination targeting lower ination? If yes, then when? Inter
national Journal of Monetary Economics and Finance, 10(3/4), 379403.
Hsiao, C. (2003). Analysis of Panel Data (2ª ed.). New York: Cambridge University Press.
Hsiao, C. (2007). Panel data analysis—advantages and challenges. TEST, 16, 1–22.
Hsiao, C. (2014). Analysis of Panel Data (3ª ed.). Cambridge: Cambridge University
Press.
Estudios económicos N° 76, Enero  Junio 2020. 147195 193
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Im, K., Pesaran, M., & Shin, Y. (2003). Testing for unit roots in heterogeneous pan
els. Journal of Econometrics, 115(1), 5374.
Jha, C. (2019). Financial reforms and corruption: Evidence using GMM estimation.
International Review of Economics & Finance, 62, 6678.
Kancs, D., & Siliverstovs, B. (2016). R&D and nonlinear productivity growth.
Research Policy, 45(6), 634646.
Kang, J., & Pueger, C. (2015). Ination Risk in Corporate Bonds. e Journal of
Finance, 70(1), 115162.
Kung, H., & Schmid, L. (2015). Innovation, Growth, and Asset Prices. e Journal
of Finance, 70(3), 10011037.
Kunieda, T., Okada, K., & Shibata, A. (2016). Corruption, Financial Development
and Economic Growth: eory and Evidence From an Instrumental Vari
able Approach With Human Genetic Diversity. Economic Notes: Review of
Banking, Finance and Monetary Economics, 45(3), 353392 .
Labra, R., & Torrecillas, C. (2018). Estimating dynamic Panel data. A practical
approach to perform long panels. Revista Colombiana de Estadística, 41(1),
3152.
Lee, Y., & Phillips, P. (2015). Model selection in the presence of incidental param
eters. Journal of Econometrics, 188(2), 474489.
Lerner, J., Sorensen, M., & Strömberg, P. (2011). Private Equity and LongRun
Investment: e Case of Innovation. e Journal of Finance, 66(2), 445
477 .
Levin, A., Lin, C., & Chu, J. (2002). Unit root tests in panel data: asymptotic and
nitesample properties. Journal of Econometrics, 108(1), 124.
LópezVillavicencio, A., & Mignon, V. (2011). On the impact of ination on output
growth: Does the level of ination matter? Journal of Macroeconomics,
33(3), 455464.
Manseld, E. (1980). Research and Development, Productivity, and Ination. Sci
ence, 209(4461), 10911093.
Mehic, A. (2018). Industrial employment and income inequality: Evidence from
panel data. Structural Change and Economic Dynamics, 45(C), 8493.
Mishkin, F. (2007). Ination Dynamics. International Finance, 10(3), 317334.
Mishkin, F. (2008). Does stabilizing ination contribute to stabilizing economic activ
ity? (NBER, Working PapersNo. 13970). Recuperado de https://www.nber.
org/system/les/working_papers/w13970/w13970.pdf
Mishra, V. (2007). e Determinants of R&D Expenditure of Firms: evidence from
a cross‐section of Indian rms. Economic Papers, 26(3), 237248.
Montresor, S., & Vezzani, A. (2015). e production function of top R&D investors:
Accounting for size and sector heterogeneity with quantile estimations.
Research Policy, 44(2), 381393.
Estudios económicos N° 76, Enero  Junio 2020. 147195194
ESTUDIOS ECONOMICOS
Myles, G., & Youse, H. (2015). Corruption and Seigniorage. Journal of Public Eco
nomic eory, 17(4), 480–503 .
Neyman, J., & Scott, E. (1948). Consistent estimates based on partially consistent
observations. Econometrica, 16(1), 132.
Oikawa, K., & Ueda, K. (2015). e Optimal Ination Rate under Schumpeterian
Growth. Centre for Applied Macroeconomic Analysis (CAMA Working
Paper No. 14/2015), Canberra, Australia: Centre for Applied Macroeco
nomic Analysis.
Osabutey, E., & Okoro, C. (2015). Political Risk and Foreign Direct Investment in
Africa: e Case of the Nigerian Telecommunications Industry. under
bird International Business Review , 57(6), 417429 .
Ozdemir, M., & Tuzunturk, S. (2009). Is price stability enough? Macroeconomic
performance of ination targeting in developing countries. International
Journal of Sustainable Economy, 1(4), 352372.
Pires, A. (2009). R&D and endogenous asymmetries between rms. Economics Let
ters, 103(3), 153156.
Ramzi, T., & Viem, J. (2016). Causality Nexus between Economic Growth, Ination
and Innovation. Journal of the Knowledge Economy, 7(1),124.
Rocha, L., Cardenas, L., Lopes, F., Oliveira, F., & Fernandes, K. (2018). e Impact of
R&D Investments on Performance of Firms in Dierent Degrees of Prox
imity to the Technological Frontier. Economics Bulletin, 38(2), 11561170.
Rocha, L., Lima, P., Khan, A., & Sousa, E. (2019). R&D Spillovers, Innovation and
Market Value: evidence of absorptive capacity in the generation of clean
technologies. Estudios de Economía Aplicada, 37(2), 182195.
Roodman, D. (2009). A note on the theme of too many instruments. Oxford Bulletin
of Economics and Statistics, 71(1), 135158.
Solow, R. (1956). A Contribution to the eory of Economic Growth. e Quarterly
Journal of Economics, 70(1), 6594.
Stein, J. (2012). Monetary Policy as Financial Stability Regulation. e Quarterly
Journal of Economics , 127(1), 5795.
Stockemer, D. (2013). Corruption and Turnout in Presidential Elections: A Mac
roLevel Quantitative Analysis. Politics & Policy , 41(2), 189212 .
Taylor, J. (2000). Teaching Modern Macroeconomics at the Principles Level. e
American Economic Review, 90(2), 9094.
Umar, M., Dahalan, J., & Aziz, M. (2016). e South African monetary policy and
ination targeting as a nominal anchor: Does the monetary policy become
more eective? International Journal of Monetary Economics and Finance,
9(4), 401  416.
Wash, C. (2003). Accountability, Transparency, and Ination Targeting. Journal of
Money, Credit and Banking, 35(5), 829849.
Estudios económicos N° 76, Enero  Junio 2020. 147195 195
INFLATION AND INNOVATION VALUE: HOW INFLATION AFFECTS INNOVATION...
Were, M., & Wambua, J. (2014). What factors drive interest rate spread of com
mercial banks? Empirical evidence from Kenya. Review of Development
Finance, 4(2), 7382.
Wu, H. (2010). Distance to frontier, intellectual property rights, and economic
growth. Economics of Innovation and New Technology, 19(2), 165183.
Wu, Y., & Zhang, J. (1998). Endogenous growth and the welfare costs of ination:
a reconsideration. Journal of Economic Dynamics and Control, 22(3), 465
482.
Wynne, M. (2008). Core ination: are view of some conceptual issues. Federal Reserve
Bank of St. Louis Review, 90(3, Part 2), 205228.
© 2021 por los autores; licencia no exclusiva otorgada a la revista Estudios eco
nómicos. Este artículo es de acceso abierto y distribuido bajo los términos y con
diciones de una licencia AtribuciónNo Comercial 4.0 Internacional (CC BYNC
4.0) de Creative Commons. Para ver una copia de esta licencia, visite http://crea
tivecommons.org/licenses/bync/4.0