The current issue and full text archive of this journal is available on Emerald Insight at:

https://www.emerald.com/insight/0144-3585.htm

Effects of economic policy Effects of

economic
uncertainty and political policy
uncertainty
uncertainty on business confidence
and investment 577
Gabriel Caldas Montes and Fabiana da Silva Leite Nogueira Received 2 December 2020
Fluminense Federal University, Niter
oi, Brazil Revised 20 February 2021
2 April 2021
Accepted 9 April 2021

Abstract
Purpose – This study estimates the effects of political uncertainty and economic policy uncertainty on
business confidence. Moreover, it also examines business confidence as a transmission channel of political
uncertainty and economic policy uncertainty to investment.
Design/methodology/approach – The study addresses the Brazilian case from May 2004 to December 2017.
Brazil experienced situations of political instability and public distrust in government and its policies, which
reflected on the economic environment. The study uses two business confidence indicators that capture
entrepreneurs’ sentiment in relation to their business and the economy. All models are estimated using ordinary
least squares and generalized method of moments.
Findings – The estimates reveal that increases in both political uncertainty and economic policy uncertainty
reduce business confidence. The findings also indicate that business confidence acts as a transmission
mechanism, i.e. uncertainties affect investments through business confidence.
Practical implications – The findings point to the following practical implications related to the existence of
uncertainties in the Brazilian economy: different institutional difficulties and government indecisions have
blurred the political scene and caused political uncertainties. In addition, the same aspects that blurred the
political scene also caused uncertainties in relation to economic policy that undermined business confidence,
and affected investment.
Originality/value – There is a vast literature on business confidence, as well as studies addressing the
relationship between business confidence and investment. This study differs from other studies as follows: in
addition to the political uncertainty, it also analyzes the effect of economic policy uncertainty on business
confidence; it uses different measures to capture political instability, and it analyzes whether business
confidence acts as a transmission channel of both uncertainties to investments.
Keywords Investment, Business confidence, Economic policy uncertainty, Political uncertainty
Paper type Research paper

1. Introduction
The literature on business confidence is vast. If on the one hand some studies indicate that
business confidence acts as a leading indicator of macroeconomic activity and influences the
economic environment, on the other hand, some studies investigate the determinants of
business confidence (Khan and Upadhayaya, 2020).
Although many advances have been made, the literature on the determinants of business
confidence continues to evolve. Some studies analyze not only the effects of macroeconomic
variables, but also the effects of other variables able to create (or reduce) uncertainties, such
as corruption (Montes and Almeida, 2017) and monetary policy credibility (Montes, 2013;
de Mendonça and Almeida, 2019). These studies reveal that low credibility and high levels of
corruption reduce confidence due to the uncertainties that emerge.
Uncertain economic scenarios created by economic policy uncertainty undermine
confidence, and affect the decision making of entrepreneurs, who, for example, postpone Journal of Economic Studies
investment and employment decisions in order to gain more information (Bloom et al., 2018). Vol. 49 No. 4, 2022
pp. 577-602
Regarding the definition of economic policy uncertainty, Al-Thaqeb and Algharabali (2019) © Emerald Publishing Limited
0144-3585
points out that “Policy uncertainty is the economic risk associated with undefined future DOI 10.1108/JES-12-2020-0582
JES government policies and regulatory frameworks” (Al-Thaqeb and Algharabali, 2019, p. 2).
49,4 Baker et al. (2016) and Al-Thaqeb and Algharabali (2019) suggest that economic policy
uncertainty delay economic recoveries during periods of recession as businesses and
households postpone their decisions about investment and consumption expenditures due to
market uncertainty. Nevertheless, regarding the effects of economic policy uncertainty on
research and development (R&D) expenditures and innovation outputs, Tajaddini and
Gholipour (2020) find positive relationships for a set of 19 developed and developing
578 countries, thus, contradicting those that claim a negative association between economic
policy uncertainty and R&D expenditure.
Since the work of Bloom (2009), and due to existing controversies in the literature, studies
investigate the effects of uncertainty shocks on different economic variables (e.g., Baker et al.,
2016; Bachmann et al., 2013; Colombo, 2013; Nodari, 2014; Donadelli, 2015; Gulen and Ion,
2015; Moore, 2017; Istiak and Serletis, 2018; Bahmani-Oskooee and Nayeri, 2018; Bahmani-
Oskooee et al., 2018; Mumtaz and Surico, 2018; Gholipour, 2019; Greenland et al., 2019; Istiak
and Alam, 2019, 2020; Tajaddini and Gholipour, 2020). In general, the findings suggest that
macroeconomic variables such as GDP, investment and employment are adversely affected
by increased economic policy uncertainty.
The political environment is also a source of uncertainty that affects the economy. Studies
provide evidence that the instability of the political environment has negative effects on the
economic environment (e.g., Barro, 1991; Alesina and Perotti, 1996; Svensson, 1998;
Carmignani, 2003; Aisen and Veiga, 2006, 2013; Durnev, 2010; Zouhaier and Kefi, 2012; Julio
and Yook, 2012; Uddin et al., 2017; Azzimonti, 2018; Jens, 2017). These studies show that
political instability has negative effects on inflation, GDP and unemployment.
Political uncertainty reflects instabilities on the political scene (i.e., involving politicians).
The instabilities arising from the political scenario are associated to uncertainties regarding
possible changes in the “rules of the game” and in the functioning of institutions. Hence, the
uncertainty related to the political system is a key feature affecting the business environment,
which entrepreneurs must consider when deciding, for instance, to start or expand their
businesses. The effects of political uncertainty are stronger when firms and politicians have
close connections and political favors might be at play.
One can suggest economic policy uncertainty reduces entrepreneurs’ optimism about the
future of the economy and their business. Similarly, an uncertain political environment can
deteriorate business confidence, producing negative effects on the economic environment.
Hence, some important questions arise. Does political uncertainty affect business confidence?
Is business confidence affected by economic policy uncertainty? Are political uncertainty and
economic policy uncertainty transmitted to investment decisions through business
confidence? These questions are particularly important for developing countries since
these countries often present higher levels of political uncertainty and economic policy
uncertainty.
This study estimates the effects of both political uncertainty and economic policy
uncertainty on business confidence in Brazil. The study also examines business confidence as
a transmission channel of political uncertainty and economic policy uncertainty to
investment. Therefore, to show the effects of both uncertainties on investment through
business confidence, first, we estimate the effect of both political uncertainty and economic
policy uncertainty on business confidence, and then we estimate the effect of business
confidence on investment.
In Brazil, political instability caused uncertainties in the economy, as institutional reform
agendas were paralyzed due to the postponement of votes in the Congress and the Senate. In
Brazil, 2015 was marked by the difficulty of articulation between the Federal Government
and the National Congress and by the postponement of important institutional reforms (such
as that of social security). After the political crisis was installed and the economic slowdown
was observed, business confidence decreased significantly due to uncertainties related to the Effects of
political environment and the conduct of economic policies. This is the logic that surrounds economic
the market in times of uncertainty, and this is directly reflected in investment decisions. This
is because in times of uncertainty, entrepreneurs take more conservative positions and
policy
postpone their expansion plans, transferring investments to options with greater liquidity. uncertainty
Brazilian entrepreneurs pointed out that the political crisis and the uncertainties regarding
the conduct of economic policies reduced the pace of business, production and job creation [1].
In Brazil, the 2015/2016 biennium is one of the most critical of the Brazilian economy. The 579
GDP fell for two consecutive years, with an accumulated decline of 7.2%.
Thus, Brazil is an interesting case study because it is an important developing country
that experienced situations of political instability and public distrust regarding the
government and economic policies, which reflected on the economic environment. Since 2015,
when Brazil had its sovereign rating downgraded and lost the investment grade seal, the
country has been experiencing loss of fiscal credibility (Montes and Costa, 2020), political
uncertainties, and increases in uncertainties about the conduct of economic policies.
This study differs from other studies in different aspects. For example, while de Mendonça
and Almeida (2019) use a Brazilian government negative assessment measure as a proxy for
political instability, our study uses different measures to capture political instability, and not
just a negative assessment of the Brazilian government [2]. Besides, we analyze the effect of
economic policy uncertainty on business confidence, and we investigate whether business
confidence acts as a transmission channel of both political uncertainty and economic policy
uncertainty to investment decisions.
The findings reveal that increases in both political uncertainty and economic policy
uncertainty adversely affect business confidence. The findings also indicate that these
uncertainties affect investments through business confidence. Therefore, regarding the
Brazilian case, the study brings the following practical implications related to the existence of
uncertainties: the difficulty of approving institutional adjustment measures, indecision about
whether or not president Dilma would remain in office, the uncertainty about who would be
the new president, who would form the new economic team and, whether the new president
would have the political capacity to implement the necessary adjustment programs have
blurred the political scene and caused political uncertainties. In addition, the same aspects
also generated uncertainties in relation to economic policy that undermined business
confidence, and affected investment. If on the one hand uncertainty destroys business
confidence, on the other hand, the indications that there will be greater political stability and
more clarity in relation to the conduct of economic policies strengthen the confidence of
entrepreneurs and foster the intention to resume the investment that was paralyzed.

2. Business confidence, uncertainties and investments

Given the importance of business confidence to economic decisions, some studies analyze the
determinants of business confidence. Konstantinou and Tagkalakis (2011) found that tax
cuts increase consumer and business confidence, while very high public debt as a proportion
of GDP reduces consumer and business confidence. Montes and Bastos (2013) analyze the role
of macroeconomic variables on business confidence, and provide evidence that business
confidence is influenced by these variables.
In addition to macroeconomic variables, other variables were also analyzed. Montes
(2013), Montes and Bastos (2013) and de Mendonça and Almeida (2019) show that greater
monetary policy credibility increases business confidence. In turn, Montes and Almeida
(2017) show that corruption adversely affects business confidence. The findings of Mart
ınez-
Serna and Navarro (2015) suggest increased interest rate volatility deteriorates the
expectations of economic agents.
JES Although the literature is growing, some gaps still exist. For instance: what are the effects
49,4 of political uncertainty and economic policy uncertainty on business confidence?
Studies addressing the consequences of political instability (uncertainty) reveal negative
effects of unstable environments on economic growth and capital accumulation (e.g., Barro,
1991; Svensson, 1998; Uddin et al., 2017). Aisen and Veiga (2013), in a panel of 169 countries,
show that political instability adversely affects economic growth and the accumulation of
physical and human capital. Aisen and Veiga (2006) reveal negative effects of an unstable
580 political environment on inflation and seigniorage. In turn, some studies show that political
instability reduces investment (e.g., Alesina and Perotti, 1996; Durnev, 2010; Zouhaier and
Kefi, 2012; Julio and Yook, 2012; Azzimonti, 2018; Jens, 2017) and foreign direct investment
(e.g., Julio and Yook, 2016).
Bloom (2009) provides evidence that increases in political and economic uncertainties
affect investment and employment. In line with Bloom (2009), other studies point to the
adverse effects of unexpected increases in uncertainty on output, employment, productivity,
consumption and investment (e.g., Colombo, 2013; Caggiano et al., 2014; Nodari, 2014;
Alexopoulos and Cohen, 2015; Leduc and Liu, 2016). Using the Economic Policy Uncertainty
€
index (EPU) proposed by Baker et al. (2016), Nodari (2014) and Stockhammar and Osterholm
(2016) show that macroeconomic variables such as GDP and employment are negatively
affected by increased economic policy uncertainty.
Studies indicate that firms adopt more conservative policies during times of high
economic policy uncertainty when the cost of borrowing increases (P
astor and Veronesi,
2012; Kelly et al., 2016; Jens, 2017; Colak et al., 2017; Al-Thaqeb and Algharabali, 2019). As
surveyed by Al-Thaqeb and Algharabali (2019), during times of high economic policy
uncertainty, firms spend less on capital (Gulen and Ion, 2015), launch fewer initial public
offerings (IPOs) (Colak et al., 2017), engage in fewer merger and acquisition (M&A) activities
(Bonaime et al., 2018), and hold more cash (Demir and Ersan, 2017; Im et al., 2017).
Regarding business and consumer confidence, few studies address the effects of political
uncertainty and economic policy uncertainty. Donadelli (2015) proposes three measures of
economic uncertainty based on Google searches. The results indicate that measures of economic
policy uncertainty in the US affect consumer sentiment and other macroeconomic variables. Also,
for the US economy, Mumtaz and Surico (2018) suggest shocks of uncertainty affect the real
economy as well as consumer and business confidence – their results indicate that uncertainty
about public debt has a large and persistent effect on consumer and business confidence, and has
negative effects on output, consumption and investment. The work of de Mendonça and Almeida
(2019) finds that political uncertainty affects business confidence in Brazil.
In turn, there exist studies addressing the relationship between business confidence and
investment (e.g., Khan and Upadhayaya, 2020), and whether business confidence acts as a
transmission channel (e.g., Montes, 2013; Montes and Bastos, 2013).
Khan and Upadhayaya (2020) analyze whether business confidence matter for
investment. They find that business confidence leads US business investment growth by
one-quarter; and business confidence has predictive ability for investment growth. Montes
(2013) finds evidence that entrepreneurs’ expectations act as a monetary policy transmission
channel to investment decisions. Montes and Bastos (2013) reveal that entrepreneurs’
expectations and business confidence act as monetary and fiscal policies transmission
channels to industrial production.
We contribute to the literature by assessing whether business confidence works as a
transmission channel of political uncertainty and economic policy uncertainty to investment.

3. Data and methodology

Based on the argument that trust in the government can affect the risk perception of
entrepreneurs (OECD, 2013), our study examines whether political uncertainty and economic
policy uncertainty affects business confidence, and as a consequence, investments. Figure 1 Effects of
describes the effects of political uncertainty and economic policy uncertainty on business economic
confidence, the effect of business confidence on investments, and thus the indirect effect of
uncertainties on investments through business confidence. The study addresses the Brazilian
policy
case and covers the period from May 2004 to December 2017. uncertainty
Regarding business confidence, we use two indicators that capture entrepreneurs’
sentiment in relation to their business and the future of the economy. The first is the Industry
Confidence Index (ICI) prepared by Get
ulio Vargas Foundation (FGV). This index synthesizes 581
a broad survey of manufacturing industry entrepreneurs, and consists of the following main
aspects: total demand level (internal and external), inventory level, current business situation,
and expectations about production and employment. Values above 100 indicate optimism of
entrepreneurs about the future of business and the economy.
As robustness, we also use the Business Confidence Index (BCI) prepared by the
Organization for Economic Cooperation and Development (OECD). The BCI provides
information on future developments, based upon opinion surveys on developments in
production, orders and stocks of finished goods in the industry sector. Numbers above 100
suggest an increased confidence in near future business performance, and numbers below
100 indicate pessimism towards future performance.
Figure 2 shows the paths of the standardized [3] series of BCI and ICI. One can observe
periods of declining business confidence, such as in the Global Financial Crisis (GFC) in 2008.
Besides, according to de Mendonça and Almeida (2019), since 2014, irresponsible policies and
political scandals in the government of President Dilma Rousseff caused deterioration of
business confidence.
To represent political instability/uncertainty, we calculated a “Political Uncertainty”
variable (PU), which is obtained from the extraction of the first principal component of three
indicators of “negative” government assessment. The three indicators are produced by the
National Confederation of Industries (CNI) in partnership with IBOPE from opinion polls
about the federal government and its policies. The nationwide survey of voters from the age
of 16 seeks to identify public perceptions of the “Way of Governing,” “Trust in the President”
and “Evaluation of the Government.” In this study, we consider only the series that capture
negative perspectives on the Government, as a proxy for political uncertainty.
Figure 3 shows the graphs of the three series that capture the negative government
assessment and the first principal component of the three series, which forms the Political
Uncertainty indicator (PU). “Poor or very poor Evaluation of the Government” expresses the
percentage of voters polled who rate the government as poor or very poor (in blue). “Do not
Trust in the President” expresses the percentage of voters polled who do not trust in the
President (in red). “Disapproves of the way of governing” expresses the percentage of
respondents who disapprove the president’s way of governing (in green). From March 2013,
distrust in relation to the government has risen sharply, coinciding with the fact that in June
2013 there were popular demonstrations against the government in Brazil. In November 2014,
following President Dilma Rousseff’s victory in the presidential election, the government’s

Figure 1.
Effects of political
uncertainty and
economic policy
uncertainty on
business confidence,
and the effect of
business confidence on
investments
JES Business confidence indexes (ICI and BCI)
49,4 2

582 0

–1

–2

–3
Figure 2. 04 05 06 07 08 09 10 11 12 13 14 15 16 17
BCI* and ICI*
(2004M05–2017M12)
B C I* IC I*

Negative assessments Political Uncertainty indicator–PU

100 5

4
80
3

60 2

1
Figure 3. 40
0
Negative government
assessments and –1
Political Uncertainty 20

indicator (PU) (2004M5 –2

– 2017M12) 0 –3
04 05 06 07 08 09 10 11 12 13 14 15 16 17 04 05 06 07 08 09 10 11 12 13 14 15 16 17

disapproval takes a new leap. In 2015, President Dilma Rousseff’s impeachment process
begins, which ends in June 2016, culminating in the beginning of Vice-President Michel
Temer’s mandate and the resumption of increased distrust in relation to the government.
Regarding economic policy uncertainty, we use the “Economic Policy Uncertainty” index
(EPU) built by Baker et al. (2016). The EPU consists of three types of components. One
component quantifies newspaper coverage of policy-related economic uncertainty. A second
component reflects the number of federal tax code provisions set to expire in future years.
The third component uses disagreement among economic forecasters as a proxy for
uncertainty. For Brazil, text archives for the newspaper Folha de S~ao Paulo from 1991
onwards are used. In each month, it is counted the number of articles containing the terms
“incerto” (uncertain) or “incerteza” (uncertainty), “econ^omico” (economic) or “economia”
(economy), and one or more of the following policy-relevant terms: “regulaç~ao, d
eficit,
orçamento, imposto, banco central, alvorada, planalto, congresso, senado, c^amara dos
deputados, legislaç~ao, lei, tarifa” (regulation, deficit, budget, tax, central bank, alvorada,
planalto, congress, senate, chamber of deputies, legislation, law, tariff). To obtain the EPU
rate, it is scaled the raw EPU counts by the number of all articles in the same newspaper and Effects of
month [4]. economic
Figure 4 presents the series for the EPU from January 2004 to December 2017. Analyzing
Figures 3 and 4, it is possible to observe an upward movement in both political uncertainty
policy
(PU) and economic policy uncertainty (EPU) in Brazil from mid-2013. uncertainty
The control variables were chosen based on the literature that addresses the determinants
of business confidence (e.g., Konstantinou and Tagkalakis, 2011; Montes, 2013; Montes and
Bastos, 2013; Montes and Almeida, 2017; de Mendonça and Almeida, 2019). Thus, the control 583
variables are: output gap [5] (Gap), expected growth rate of industrial production [6]
ðEIND_GDP Þ, inflation rate [7] ðIpcaÞ, expected inflation rate [8] ðEIpcaÞ, monetary policy
interest rate [9] ðSelicÞ, exchange rate [10] ðExchangeÞ, primary surplus [11] ðSuprimÞ,
infrastructure [12] ðInfrÞ, Credit Default Swaps [13] (CDS) as a measure of sovereign risk,
monetary policy credibility [14] (ICM), and subprime crisis [15] ðSubprimeÞ.
The use of time series data in estimations entails checking whether the series have a unit
root to avoid spurious regressions – for more details, see, Granger and Newbold (1974),
Phillips (1986) and Ventosa-Santaularia (2009). So, we did the following unit root tests:
Augmented Dickey-Fuller (ADF), Phillips-Perron (PP) and Kwiatkowski-Phillips-Schmidt-
Shin (KPSS) – Table A1 in the Appendix. The results show the following series are I(1): Ipca,
EIpca, Selic, Exchange, Suprim, CDS, ICM and PU. Thus, in the estimates, we use the first
differences of these series – i.e. DðIpcaÞ, DðEIpcaÞ, DðSelicÞ, D(Exchange), DðSuprimÞ,
DðCDSÞ, D(ICM) and D(PU).
In order to verify whether political uncertainty and economic policy uncertainty affect
business confidence, we estimate the following equations using the ICI and BCI as dependent
variables:
ICIt ¼ α0 þ α1 X0 þ α2 DðPUÞt þ ε1t (1)
ICIt ¼ α3 þ α4 X0 þ α5 EPUt þ ε2t (2)
ICIt ¼ α6 þ α7 X0 þ α8 DðPUÞt þ α9 EPUt þ ε3t (3)
ICIt ¼ α10 þ α11 X1 þ α12 DðPUÞt þ α13 EPUt þ ε3t (4)

Economic Policy Uncertainty index (EPU)

700

600

500

400

300

200
Figure 4.
100 Economic Policy
Uncertainty index
(EPU) (2004M5 –
0 2017M12)
04 05 06 07 08 09 10 11 12 13 14 15 16 17
JES BCIt ¼ β0 þ β1 X0 þ β2 DðPUÞt þ ξ1t (5)
49,4 BCIt ¼ β3 þ β4 X0 þ β5 EPUt þ ξ2t (6)
BCIt ¼ β6 þ β7 X0 þ β8 DðPUÞt þ β9 EPUt þ ξ3t (7)
BCIt ¼ β10 þ β11 X1 þ β12 DðPUÞt þ β13 EPUt þ ξ3t (8)

584 where, α0 ; α3 ; α6 ; α10 ; β0 ; β3 ; β6 and β10 are intercepts; in turn, α1 ; α4 ; α7 ; α11 ;

β1 ; β4 ; β7 and β11 are the vectors of parameters associated with the control variables.; εt
and ξt are the error terms. X1 is a vector of control variables formed by: Gapt, EIND_GDPt4,
D(Ipca)t, D(EIpca)t1, D(Selic)t2, D(Exchange)t, Infrt, D(Suprim)t1, D(CDS)t3, D(ICM)t1 and
Subprime [16]. The difference between X0 and X1 is that X0 does not use D(ICM)t1. Thus,
equations (3) and (4), as well as equations (7) and (8) are different due to the use of the
monetary policy credibility index as a control variable. We added the ICM to verify whether
the effects of both uncertainties would remain after controlling for this variable.
After estimates for business confidence, we analyze the influence of this variable on
investment, and thus identify the transmission of political and economic policy uncertainties
to investment through the expectations channel. Regarding the investment variable
(INVEST) [17], we follow Acosta and Loza (2005), de Mendonça and Lima (2011),
de Mendonça and Cacicedo (2015) and Degiannakis et al. (2017) and use the gross fixed
capital formation series. Figure 5 presents the graph related to INVEST.
In relation to the investment equation, the control variables were chosen based on
previous studies on the determinants of investments (e.g. Ghura and Goodwin, 2000; Acosta
and Loza, 2005; de Mendonça and Lima, 2011). Thus, the control variables are: real interest
rate [18] ðIR_RealÞ, output gap ðGapÞ, expected GDP growth [19] ðE_GDPÞ, expected inflation
rate ðEIpcaÞ, inflation rate (Ipca), stock market index [20] (Ibovespa), exchange rate
ðExchangeÞ, expected exchange rate [21] ðE_ExchangeÞ, credit as a proportion of GDP [22]
(Credit) and public debt [23] ðDebtÞ.

INVEST
40

30

20

10

–10
Figure 5.
Gross fixed capital
formation (2004M1– –20
2017M12) 04 05 06 07 08 09 10 11 12 13 14 15 16 17
 Based on the unit root tests (Table A1, Appendix), we identified that the following series Effects of
are I(1): IR_Real, E_GDP, EIpca, Ipca, Ibovespa, Exchange, E Exchange, Credit and Debt. economic
Thus, in the estimations, we use the first difference of these series, which are denoted by
DðIR_RealÞ, DðE GDPÞ, DðEIpcaÞ, D(Ipca), D(Ibovespa), D(Exchange), DðE ExchangeÞ),
policy
D(Credit) and DðDebtÞ. uncertainty
To verify if business confidence acts as a transmission channel to investment (INVEST),
the following equations are estimated:
585
INVESTt ¼ γ 0 þ γ 1 Z þ γ 2 ICIt−3 þ υ1t (9)
INVESTt ¼ γ 3 þ γ 4 Z þ γ 5 BCIt−3 þ υ2t (10)

where, γ 0 and γ 3 ; are intercepts; γ 1 and γ 4 are the vectors of parameters associated with the
control variables; γ 2 and γ 5 are the coefficients associated to business confidence (ICI and
BCI); υ1t and υ2t are the error terms. Z is a vector of control variables formed by the following
variables: DðIR_RealÞt−4, Gapt, DðE GDPÞt−4, DðEIpcaÞt−4, DðIpcaÞt−3, DðIbovespaÞt−4 ,
DðExchangeÞt−4, DðE_ExchangeÞt−4, DðCreditÞt−6, DðDebtÞt−3 [24].
Table 1 shows the correlations between the main variables used in the study. Both
business confidence indexes are negatively correlated with the uncertainty indicators (EPU,
PU and D(PU)). The business confidence indexes present a higher correlation with the
political uncertainty (PU) than with the economic policy uncertainty (EPU). The correlations
between the confidence indexes and D(PU) are smaller, however, they remain negative. As
expected, we observe a positive correlation between business confidence (ICI and BCI) and
investments (INVEST). We also observe a positive correlation (equal to 0.63) between PU and
EPU, and a positive correlation (equal to 0.14) between D(PU) and EPU.
All equations are estimated using the following methods: ordinary least squares (OLS),
generalized method of moments (GMM) with Newey–West covariance matrix (Newey and
West, 1987), and two-step generalized method of moments (GMM-2) with Windmeijer (2005)
covariance matrix. [25] GMM provides consistent estimates (Wooldridge, 2001; Hall, 2005)
and allows one to verify whether the results obtained by OLS are preserved. GMM is used to
deal with endogeneity problems. The endogeneity problem occurs (in general) due to the
omission of variables, simultaneity and measurement errors. Our analysis might be subject to
the omission of variables, simultaneity and measurement errors. When a variable is omitted
from the model, it ends up being incorporated into the error. If this omitted variable is
correlated with other regressors (which is not uncommon), then there will be a correlation
between the explanatory variable and the error term. Since not all determinants of business
confidence are known and measured, the omitted variable problem can affect the model. In
relation to the simultaneity problem, its cause is because some explanatory variable is
determined simultaneously with the dependent variable (which, in our model, is the case of
the output gap).
In order to deal with such problems, we follow the methodology of Johnston (1984) to select
the instruments on GMM estimation, i.e. the instruments were dated to the period t1 or

Variables ICI BCI INVEST EPU PU D(PU)

ICI 1.00 0.98 0.85 0.56 0.66 0.11

BCI 0.98 1.00 0.87 0.58 0.69 0.11
INVEST 0.85 0.87 1.00 0.46 0.60 0.15
EPU 0.56 0.58 0.46 1.00 0.63 0.14
PU 0.66 0.69 0.60 0.63 1.00 0.18 Table 1.
D(PU) 0.11 0.11 0.15 0.14 0.18 1.00 Correlations
JES earlier to assure the exogeneity [26]. Overidentification has an important role in the selection
49,4 of instrumental variables to improve the efficiency of the estimators (Cragg, 1983). Therefore,
a standard J-test was performed to indicate whether the orthogonality condition is satisfied.
Finally, to analyze the endogeneity of the equation regressors, we report the results of the
Durbin–Wu–Hausman test.

586 4. Results
Tables 2 and 3 present the results where the ici* and the bci* are the dependent variables,
respectively. The findings indicate that both ici* and bci* are negatively affected by political
uncertainty (PU) and economic policy uncertainty (EPU). The findings indicate that
entrepreneurs become more pessimistic about the future of the economy and their business in
a more uncertain political environment and when uncertainties related to the conduct of
economic policies increase. Based on the averages of the statistically significant coefficients
obtained for D(PU) and EPU, we observed that positive one standard deviation shocks in
these variables (indicating increases in uncertainties) cause reductions in the ici* of 0.182
basis points (bp) and 0.173 bp, respectively, and cause reductions in bci* of 0.207 bp and
0.424 bp, respectively.
Regarding the control variables, the results indicate that an increase in economic activity
ðGapÞ has a positive effect on business confidence, the coefficients are positive and
statistically significant in all estimates. Regarding Eind_GDP, estimates indicate that an
improvement in market expectations about industrial production positively influences
business confidence. In turn, the estimates for Infr suggest an improvement in the country’s
infrastructure makes entrepreneurs more optimistic about the future of the economy and
their business. In relation to the impact of the government’s fiscal effort, the results for
Suprim indicate that an increase in the government’s fiscal effort increases business
confidence.
Analyzing the results obtained for DðSelicÞ and DðExchangeÞ on ici* and bci*, we observe
negative and significant coefficients in all cases. Therefore, when DðSelicÞ increases, business
confidence reduces. Regarding DðExchangeÞ, a possible interpretation for this result is that
exchange rate devaluations increase the costs of importing raw materials and machinery and
equipment, making entrepreneurs more pessimistic. These results corroborate Montes and
Bastos (2013).
In relation to DðCDSÞ, all coefficients are negative significant. Thus, an increase in
sovereign risk reduces business confidence. The coefficients for D(Ipca) and D(EIpca) do not
show statistical significance. On the other hand, the coefficients for the effect of monetary
policy credibility (D(ICM)) are positive and significant in all estimates, reinforcing the results
obtained by de Mendonça and Almeida (2019). This is probably because Brazil adopts the
inflation targeting regime and, therefore, what matters for business confidence is whether the
established inflation target can anchor expectations. Finally, Subprime does not present
statistical significance.
Table 4 presents the results of equations (9) and (10). The findings suggest business
confidence affects investment (INVEST). All coefficients for ici* and bci* are positive and
significant. Thus, increases in political uncertainty and economic policy uncertainty
negatively influence investment through the expectations channel (business confidence). The
findings corroborate recent results provided by the empirical literature addressing whether
business confidence predicts investment (e.g., Khan and Upadhayaya, 2020).
The results obtained for DðIR_RealÞ suggest a tight monetary policy reduces the level of
investment. This result corroborates de Mendonça and Lima (2011) and Montes (2013).
The estimates for both Gap and E_GDP reveal that when economic activity and
expectations for the GDP growth increase, investments also increase.
Estimator OLS GMM GMM-2
Variables (1) (2) (3) (4) (1) (2) (3) (4) (1) (2) (3) (4)

Constat 4.4661*** 3.5017*** 3.4770*** 3.6926*** 4.4891*** 5.1228*** 4.0705*** 5.1450*** 4.3850*** 3.7467** 3.9979** 4.6333***
(1.1479) (1.1619) (1.1621) (1.2039) (1.1766) (0.9639) (1.1315) (0.9679) (1.4736) (1.6852) (1.6177) (1.4715)
Gapt 0.0696* 0.0865** 0.0868** 0.0882** 0.1259*** 0.1050*** 0.1359*** 0.1389*** 0.1078*** 0.1336*** 0.1339*** 0.1357***
(0.0376) (0.0371) (0.0369) (0.0353) (0.0281) (0.0239) (0.0310) (0.0261) (0.0406) (0.0436) (0.0307) (0.0367)
EIND_GDPt4 0.2054*** 0.1590*** 0.1614*** 0.1635*** 0.1561*** 0.1246*** 0.1276*** 0.1162*** 0.1573*** 0.1189** 0.1411*** 0.1549***
(0.0371) (0.0494) (0.0501) (0.0481) (0.0235) (0.0308) (0.0308) (0.0324) (0.0407) (0.0505) (0.0496) (0.0401)
D(Ipca)t 0.2751* 0.1819 0.1719 0.2315* 0.5064* 0.2729 0.1137 0.3175 0.6140 0.2048 0.1393 0.2453
(0.1491) (0.1417) (0.1413) (0.1366) (0.2925) (0.2449) (0.3530) (0.3651) (0.5390) (0.4132) (0.3703) (0.2164)
D(EIpca)t1 0.1224 0.0216 0.0439 0.0964 0.2242 0.3447 0.1749 0.0240 0.0520 0.1527 0.0485 0.1369
(0.1498) (0.1393) (0.1403) (0.1454) (0.2831) (0.2848) (0.3521) (0.3740) (0.5883) (0.4545) (0.3598) (0.3627)
D(Selic)t2 0.5286*** 0.5619*** 0.5337*** 0.5639*** 0.6371*** 0.6783*** 0.5144** 0.6616*** 0.6080* 0.6183** 0.5031* 0.5335***
(0.1806) (0.1644) (0.1631) (0.1659) (0.9408) (0.2021) (0.2486) (0.2489) (0.3291) (0.2757) (0.2721) (0.1787)
D(Exchange)t 1.1592*** 1.0882*** 0.9774** 1.1025*** 1.6237* 2.3950*** 1.7777** 2.6961*** 2.2374* 2.3061** 1.7579* 1.8226**
(0.4207) (0.3589) (0.3900) (0.3986) (0.9408) (0.7726) (0.8412) (0.8143) (1.2034) (1.0248) (1.0253) (0.8574)
D(Suprim)t1 0.5731** 0.6371** 0.6378** 0.6675** 2.8094*** 2.1256*** 2.3817*** 2.4597*** 2.6671** 1.7521** 1.3974* 0.9507***
(0.2477) (0.2649) (0.2594) (0.2642) (0.5250) (0.4119) (0.5972) (0.5239) (1.2926) (0.6765) (0.7115) (0.3627)
Infrt 0.2588*** 0.2236*** 0.2207*** 0.2351*** 0.2747*** 0.3359*** 0.2640*** 0.3482*** 0.26317** 0.2453** 0.2587** 0.2948***
(0.0817) (0.0756) (0.0761) (0.0801) (0.0834) (0.0601) (0.0758) (0.0650) (0.1053) (0.1160) (0.1143) (0.0987)
D(CDS)t3 0.0038** 0.0036** 0.0035** 0.0036** 0.0081*** 0.0111*** 0.0092*** 0.0077*** 0.0110*** 0.0068*** 0.0065*** 0.0053***
(0.0016) (0.0014) (0.0014) (0.0015) (0.0021) (0.0019) (0.0016) (0.0023) (0.0029) (0.0018) (0.0019) (0.0019)
D(ICM)t1 0.9355** 4.3579*** 2.0884*
(0.4413) (1.1877) (1.1074)
D(PU)t 0.3755* 0.2747* 0.2947* 1.2805*** 1.0085*** 1.0189*** 1.1485* 0.7490* 0.7459*
(0.1976) (0.1656) (0.1723) (0.3660) (0.3775) (0.3895) (0.6704) (0.4420) (0.3890)
EPUt 0.0021** 0.0020** 0.0019** 0.0016** 0.0014** 0.0019** 0.0018* 0.0016** 0.0016*
(0.0009) (0.0009) (0.0009) (0.0007) (0.0006) (0.0007) (0.0010) (0.0007) (0.0008)
Subprime 0.4518 0.1914 0.2582 0.2737 0.9284 0.2035 0.6405 0.4031 0.5062 0.4447 0.5214 0.3851
(0.4328) (0.4206) (0.4169) (0.4017) (0.6125) (0.4799) (0.6054) (0.6549) (1.0409) (0.6829) (0.5653) (0.3909)
Adj. R2 0.65 0.67 0.67 0.67 0.27 0.49 0.41 0.38 0.28 0.57 0.61 0.66
J-statistic 21.13 25.59 0.41 21.79 22.41 26.14 26.22 24.78
Prob (J- 0.78 0.60 0.77 0.70 0.87 0.71 0.79 0.85
statistic)
D-W-H test 4.30 2.36 3.41 2.79 3.78 3.31 3.72 3.21
Prob (D-W-H) 0.96 0.99 0.99 0.99 0.97 0.98 0.98 0.99
Observations 164 164 164 164 160 160 160 160 157 157 157 155
Instruments 39 40 39 40 43 43 46 47
N Instr./N Obs 0.24 0.25 0.24 0.25 0.27 0.27 0.29 0.30
Note(s): Marginal Significance Levels: *** denotes 0.01, ** denotes 0.05 and * denotes 0.1. Standard errors are in parentheses. “Prob J-Statistic” reports the respective
p-value of the J-test. D-W-H test is the Durbin-Wu-Hausman test (difference in J-stats) and null hypothesis is that the regressors are exogenous
uncertainty
economic

587
policy
Effects of

variable: ici*)
Table 2.

(dependent
OLS, GMM and GMM-
2 estimations
 JES
49,4

588

Table 3.

(dependent
2 estimations

variable: bci*)
OLS, GMM and GMM-
Estimator OLS GMM GMM-2
Variables (5) (6) (7) (8) (5) (6) (7) (8) (5) (6) (7) (8)

Constat 3.5149*** 2.4460** 2.4223** 2.6723** 5.2175*** 4.7284*** 3.8077*** 4.1546*** 4.0176** 3.6636*** 3.0902** 3.7345**
(1.0520) (1.0655) (1.0641) (1.0868) (1.1260) (0.9657) (1.0980) (0.9391) (1.6554) (1.3649) (1.5181) (1.7158)
Gapt 0.0780** 0.0968*** 0.0971*** 0.0986*** 0.1081*** 0.1071*** 0.1313*** 0.1231*** 0.1375*** 0.1261*** 0.1479*** 0.1546***
(0.0339) (0.0324) (0.0323) (0.0305) (0.0240) (0.0233) (0.0301) (0.0254) (0.0356) (0.0395) (0.0411) (0.0350)
EIND_GDPt4 0.2142*** 0.1634*** 0.1656*** 0.1682*** 0.1600*** 0.1270*** 0.1016*** 0.0931** 0.1788*** 0.1289*** 0.1383*** 0.1561***
(0.0368) (0.0491) (0.0495) (0.0474) (0.0247) (0.0271) (0.0308) (0.0323) (0.0397) (0.0471) (0.0451) (0.0334)
D(Ipca)t 0.2390 0.1345 0.1249 0.1941 0.5080** 0.3259 0.1576 0.2987 0.2798 0.1408 0.1761 0.0231
(0.1517) (0.1488) (0.1480) (0.1448) (0.2540) (0.2135) (0.3401) (0.3514) (0.4238) (0.4609) (0.3971) (0.3206)
D(EIpca)t1 0.2213 0.1132 0.1346 0.1955 0.0700 0.2328 0.6361 0.7391 0.1948 0.3251 0.0903 0.5565
(0.1719) (0.1496) (0.1558) (0.1888) (0.2123) (0.2308) (0.4181) (0.4907) (0.4236) (0.5710) (0.4224) (0.5443)
D(Selic)t2 0.5431*** 0.5757*** 0.5487*** 0.5838*** 0.7288*** 0.7923*** 0.6012*** 0.7834*** 0.5202* 0.5413* 0.4842** 0.4708**
(0.1901) (0.1716) (0.1696) (0.1717) (0.1980) (0.1735) (0.2129) (0.2305) (0.3013) (0.2773) (0.2394) (0.2004)
D(Exchange)t 1.1503*** 1.0556*** 0.9494** 1.0945*** 2.9159*** 2.7715*** 3.1969*** 3.4819*** 3.0505** 2.5791*** 2.3826** 1.7351*
(0.4404) (0.3642) (0.4024) (0.4087) (0.8062) (0.5753) (0.9009) (0.7783) (1.4141) (0.9688) (1.1818) (0.9137)
D(Suprim)t1 0.5953** 0.6660*** 0.6667** 0.7011** 2.4984*** 1.9190*** 2.3766*** 2.6041*** 2.0727*** 1.6491** 1.1303*** 0.9501***
(0.2788) (0.2929) (0.2886) (0.2923) (0.4801) (0.4408) (0.6775) (0.6656) (0.6903) (0.8265) (0.3960) (0.3565)
Infrt 0.1921** 0.1528** 0.1501** 0.1668** 0.3237*** 0.3137*** 0.2595*** 0.2919*** 0.2366** 0.2456*** 0.1986* 0.2361**
(0.0749) (0.0684) (0.0687) (0.0714) (0.0806) (0.0628) (0.0719) (0.0631) (0.1173) (0.0894) (0.1046) (0.1141)
D(CDS)t3 0.0036** 0.0034*** 0.0033*** 0.0034** 0.0090*** 0.0105*** 0.0081*** 0.0072*** 0.0071** 0.0085*** 0.0057*** 0.0033*
(0.0014) (0.0012) (0.0012) (0.0013) (0.0021) (0.0019) (0.0018) (0.0020) (0.0033) (0.0027) (0.0018) (0.0019)
D(ICM)t1 1.0844*** 3.8109*** 2.5736**
(0.3926) (1.1003) (1.0008)
D(PU)t 0.3746* 0.2632 0.2864 1.0503*** 0.6940*** 0.7634** 1.4928** 0.7886** 0.8833**
(0.2170) (0.1772) (0.1866) (0.3677) (0.3367) (0.3808) (0.7417) (0.3958) (0.3999)
EPUt 0.0023** 0.0022** 0.0021** 0.0020*** 0.0020*** 0.0026*** 0.0026* 0.0018** 0.0018**
(0.0009) (0.0009) (0.0009) (0.0007) (0.0007) (0.0008) (0.0013) (0.0008) (0.0007)
Subprime 0.4279 0.1501 0.2141 0.2320 0.2487 0.1418 0.2935 0.3082 0.5939 0.2287 0.1820 0.3492
(0.2922) (0.2761) (0.2743) (0.2608) (0.3340) (0.3541) (0.3189) (0.3622) (0.4369) (0.5360) (0.3736) (0.3621)
2
Adj. R 0.66 0.68 0.69 0.70 0.38 0.54 0.41 0.41 0.39 0.57 0.63 0.64
J-statistic 20.54 23.14 20.41 19.69 21.93 24.05 23.05 21.42
Prob (J- 0.81 0.73 0.77 0.81 0.95 0.84 0.92 0.92
statistic)
D-W-H test 4.51 2.07 3.41 3.75 3.51 2.66 3.61 4.22
Prob (D-W-H) 0.95 0.99 0.99 0.99 0.98 0.99 0.99 0.99

(continued )
Estimator OLS GMM GMM-2
Variables (5) (6) (7) (8) (5) (6) (7) (8) (5) (6) (7) (8)

Observations 164 164 164 164 160 160 160 160 157 157 158 156
Instruments 39 40 39 40 46 44 47 46
N Instr./N Obs 0.24 0.25 0.24 0.25 0.29 0.28 0.30 0.29
Note(s): Marginal Significance Levels: *** denotes 0.01, ** denotes 0.05 and * denotes 0.1. Standard errors are in parentheses. “Prob J-Statistic” reports the respective
p-value of the J-test. D-W-H test is the Durbin-Wu-Hausman test (difference in J-stats) and null hypothesis is that the regressors are exogenous
uncertainty
economic

589
policy
Effects of

Table 3.
 JES
49,4

590

INVEST)
Table 4.
OLS, GMM and
GMM-2 estimations
(dependent variable:
Estimator OLS GMM GMM-2
Variables (9) (10) (9) (10) (9) (10)

Constat 2.9833*** (0.6065) 3.1557*** (0.5774) 2.2366** (0.4022) 2.9410*** (0.4463) 2.1589*** (0.7108) 2.2617*** (0.7877)
D(IRreal)t4 4.3956*** (1.2175) 4.2969*** (1.2584) 3.3528*** (1.1050) 3.7494*** (1.1671) 3.0906* (1.7296) 2.8757* (1.5190)
Gapt 0.7388*** (0.2476) 0.6163*** (0.2238) 0.5569*** (0.1997) 0.4185** (0.2080) 0.6409** (0.3066) 0.5498 (0.4071)
D(EGDP)t4 9.3062** (4.6701) 8.9576** (4.3318) 12.2460** (2.4188) 9.7075*** (2.6321) 10.3461** (4.7668) 8.8291* (4.6322)
D(Eipca)t4 2.9554* (1.6454) 3.3697* (1.7330) 4.1163** (2.0423) 5.0693* (2.9329) 1.3254 (4.0212) 3.0984 (1.9293)
D(Ipca)t4 3.0171** (1.2332) 2.7518** (1.2177) 2.2050* (1.2704) 2.9570** (1.4268) 3.4567* (2.0650) 3.5110* (1.9293)
D(Ibovespa)t4 0.0002* (0.0001) 0.0003** (0.0001) 0.0004** (0.0002) 0.0004** (0.0001) 0.0007** (0.0003) 0.0007** (0.0003)
D(Exchange)t4 4.9148 (3.9017) 5.2948 (3.6173) 0.42833 (6.0930) 7.3749 (5.9421) 1.8562 (10.7094) 5.4385 (11.1284)
D(Exchange)t4 0.7789 (7.8059) 0.6123 (7.2281) 3.6488 (6.4849) 0.1657 (6.5857) 8.9474 (11.9137) 6.7373 (12.0967)
D(Credit)t6 6.5468*** (1.8901) 4.8758*** (1.8141) 10.5201*** (1.4820) 5.6680*** (1.7807) 9.5157*** (2.8364) 6.4201** (2.8173)
D(Debt)t3 0.6876** (0.3116) 0.6122** (0.3033) 2.1990*** (0.4647) 1.5048*** (0.5244) 2.6017*** (0.8943) 2.1735** (1.0801)
ici*t3 6.2805*** (0.7897) 5.5190*** (0.6188) 5.2106*** (1.06299)
bci*t3 6.8116*** (0.6768) 6.6093*** (0.6473) 5.9628*** (1.1750)
Adj. R2 0.76 0.78 0.71 0.77 0.69 0.74
J-statistic 24.22 23.37 23.82 22.75
Prob (J-statistic) 0.87 0.80 0.90 0.91
D-W-H test 1.73 2.77 1.76 1.70
Prob (D-W-H) 0.99 0.99 0.99 0.99
Observations 161 161 157 159 156 156
Instruments 45 42 46 45
N Instr./N Obs 0.29 0.26 0.29 0.29
Note(s): Marginal Significance Levels: *** denotes 0.01, ** denotes 0.05 and * denotes 0.1. Standard errors are in parentheses. “Prob J-Statistic” reports the respective
p-value of the J-test. D-W-H test is the Durbin-Wu-Hausman test (difference in J-stats) and null hypothesis is that the regressors are exogenous
 The findings also indicate that accelerating inflation and rising inflation expectations Effects of
reduce investments. In the literature, these findings are associated with the adverse effect that economic
an increase in uncertainty related to the economic environment has on investment (Acosta
and Loza, 2005; de Mendonça and Lima, 2011). On the other hand, DðExchangeÞ and
policy
DðE ExchangeÞ are not statistically significant. Although exchange rate changes do not uncertainty
directly affect investments, there is an indirect effect, since exchange rate changes affect
business confidence, and the latter affects investment.
We also observe that a better stock market performance increases the level of investments, 591
corroborating Montes (2013). In turn, estimates for D(Credit) indicate a positive influence on
INVEST, reinforcing Acosta and Loza (2005) and de Mendonça and Lima (2011). The findings
for DðDebtÞ reveal negative and significant coefficients, reinforcing de Mendonça and
Lima (2011).

5. Concluding remarks
This study provided empirical evidence on the effects of political uncertainty and economic
policy uncertainty on business confidence. In addition, it investigated the role of business
confidence as a transmission channel of both uncertainties to investment. Thus, the idea was
to investigate whether these uncertainties affect business confidence, and whether business
confidence affects investments in the economy, characterizing business confidence as a
channel that transmits uncertainties to the economy.
The findings indicate that increases in political uncertainty and economic policy
uncertainty reduce business confidence. Regarding the role of confidence as a transmission
channel, the findings show that business confidence acts as a transmission channel of
political uncertainty and economic policy uncertainty to investment. Increases in political
uncertainty and economic policy uncertainty adversely influence investment through
reduced business confidence.
From the findings of this study, we can point out the following policy suggestions.
Policymakers must strengthen institutions through the development and use of rules that
prevent irresponsible actions and are capable of destabilizing the political scenario, and they
should establish management practices based on greater levels of transparency and
communication with the public in order to better guide expectations and promote an
environment that enhance business confidence. Positive evaluations about the government,
about the management of public affairs, as well as about the chief executive, tend to generate
a more optimistic business environment, positively influencing business investment
decisions. Thus, policymakers should adopt credible economic policies that promote a
more stable business environment with less uncertainty, allowing entrepreneurs to plan and
make investment decisions with a view to a longer time horizon.
Future research could analyze the effects of the uncertainties addressed in the present
study on the expectations formation process, for example, on the disagreements in
expectations about economic variables (such as, output, inflation and interest rates). Future
research could also analyze the effects of these uncertainties on the sovereign risk
represented by different variables (such as CDS spreads and sovereign credit ratings issued
by credit rating agencies).

Notes
1. Statements by Brazilian businessmen can be found at: https://epocanegocios.globo.com/360/noticia/
2017/08/como-crise-politica-afeta-os-negocios-das-maiores-empresas-do-pais.html.
2. In this study, we consider the series that capture negative perspectives on the Government, as a
proxy for political uncertainty. Thus, we use the “Way of Governing”, “Trust in the President” and
“Evaluation of the Government”
JES 3. Following de Mendonça and Almeida (2019), BCI* and ICI* are standardized series of BCI and ICI
(mean zero and standard deviation 1), respectively. In the estimations, these series will be used.
49,4
4. The data is obtained from: https://www.policyuncertainty.com/brazil_monthly.html.
5. The output gap (Gap) captures the economic activity. Real GDP is calculated using cumulative 12-
months GDP (Central Bank of Brazil (CBB), series 4,382), and the general price index (obtained from
the IPEADATA website). The economic cycle (GAP) was obtained by the Hamilton (2018) method.
592 6. Data on expectations are obtained from the CBB’s Market Expectations System, which is managed
by the Department of Investor Relations and Special Studies (GERIN). Every weekday the GERIN
collects forecasts from market participants (which can be financial institutions, real sector
companies and consulting firms). Based on market expectation provided by the CBB, we use the
expected growth rate of industrial production (EIND_GDP). EIND_GDP is the (average) market
expectation in relation to the growth of industrial production for the next 12 months.
7. Inflation is the twelve-month change in the Consumer Price Index (IPCA) (measured at a monthly
frequency) – series 13,522 provided by the CBB.
8. Based on market expectation provided by the CBB, the expected inflation rate is the (average)
market expectation in relation to the inflation rate for the next 12 months, provided by the CBB.
9. Selic interest rate (series 4,189) provided by the CBB.
10. This series is the “Exchange Rate–Free - US Dollar (Sale) - end of period” (Series 3,696), provided by
the CBB.
11. Series 5,793 provided by the CBB.
12. This series consists of the consumption of industrial electricity, which is used as a proxy for the
country’s infrastructure. The series is provided by the Energy Research Company (EPE).
13. The 5-years CDS spread is used as a proxy for sovereign risk perception (Montes and Souza (2018).
14. We use the monetary policy credibility index (ICM) developed by de Mendonça (2007).
15. Following the literature on the Brazilian economy (e.g. Montes and Almeida, 2017), this dummy
variable assumes a value equal to 1 between August 2008 and January 2009, and 0, otherwise.
16. The lags of the variables were determined based on the empirical literature, as well as following the
general-to-specific method (Hendry, 2001). Thus, the choice of the lags of the variables considers the
statistical significance of the estimated coefficients, the degree of fit of the model, and the principle
of parsimony, in order to ensure that the chosen model has explanatory power. In this sense,
regarding the lag structure of the equations employed, we tested several specifications with
different lag structures on the equation variables, and chose the one in which the lag structure was
formed by coefficients with the smallest possible lags.
17. INVEST refers to the 12-months percentage change in the Gross Fixed Capital Formation index.
The original series refers to the “IPEA indicator of gross fixed capital formation” provided by the
Instituto de Pesquisa Econ^omica Aplicada (IPEA).
18. The real interest rate is obtained by the difference between the Selic rate (series 4,189) and the
average inflation expectation measured by the IPCA–cumulative rate for the next twelve months.
Both obtained from the CBB.
19. This is the (average) market expectation in relation to the GDP growth for the next 12 months,
provided by the CBB.
20. Ibovespa (series7845), obtained from the CBB
21. This is the (average) market expectation in relation to the exchange rate for the next 12 months,
provided by the CBB.
22. This is the ratio between the balance of credit granted by the National Financial System and the
value of the accumulated GDP in the last twelve months (Series, 20,622), provided by the CBB.
23. This is the general government gross debt as a proportion of GDP (Series 4,537) provided by Effects of
the CBB.
economic
24. Once again, the lags of the variables were determined based on the empirical literature, as well as policy
following the general-to-specific method (Hendry, 2001). Particularly, with respect to the effect of
business confidence on investment, we followed the empirical literature (for instance, Khan and uncertainty
Upadhayaya (2020), which show that business confidence leads US business investment growth by
one-quarter). Thus, the choice of the lags of the variables considers the statistical significance of the
estimated coefficients, the degree of fit of the model, and the principle of parsimony, in order to 593
ensure that the chosen model has explanatory power. Thus, once again, regarding the lag structure
of the equations employed, we tested several specifications with different lag structures on the
equation variables, and chose the one in which the lag structure was formed by coefficients with the
smallest possible lags.
25. We tested autocorrelation and heteroskedasticity using the LM and ARCH tests, respectively.
26. Table A3 shows the instruments employed in the GMM analysis for each equation.

References
Acosta, P. and Loza, A. (2005), “Short and long run determinants of private investment in Argentina”,
Journal of Applied Economics, Vol. 8 No. 2, pp. 389-406.
Aisen, A. and Veiga, F.J. (2006), “Does political instability lead to higher inflation? A panel data
analysis”, Journal of Money, Credit, and Banking, Vol. 38 No. 5, pp. 1379-1389.
Aisen, A. and Veiga, F.J. (2013), “How does political instability affect economic growth?”, European
Journal of Political Economy, Vol. 29 No. 1, pp. 151-167.
Al-Thaqeb, S.A. and Algharabali, B.G. (2019), “Economic policy uncertainty: a literature review”, The
Journal of Economic Asymmetries, Vol. 20, e00133.
Alesina, A. and Perotti, R. (1996), “Income distribution, political instability, and investment”, European
Economic Review, Vol. 40 No. 6, pp. 1203-1228.
Alexopoulos, M. and Cohen, J. (2015), “The power of print: uncertainty shocks, markets, and the
economy”, International Review of Economics and Finance, Vol. 40, pp. 8-28.
Azzimonti, M. (2018), “Partisan conflict and private investment”, Journal of Monetary Economics,
Vol. 93, pp. 114-131.
Bachmann, R., Elstner, S. and Sims, E. (2013), “Uncertainty and economic activity: evidence from
business survey data”, American Economic Journal: Macroeconomics, Vol. 5 No. 2, pp. 217-249.
Bahmani-Oskooee, M. and Nayeri, M.M. (2018), “Policy uncertainty and the demand for money in
Canada: a nonlinear approach”, Applied Economics Quarterly, Vol. 64, pp. 279-295.
Bahmani-Oskooee, M., Harvey, H. and Niroomand, F. (2018), “On the impact of policy uncertainty on
oil prices: an asymmetry analysis”, International Journal of Financial Studies, Vol. 6, pp. 1-11.
Baker, S.R., Bloom, N. and Davis, S.J. (2016), “Measuring economic policy uncertainty”, Quarterly
Journal of Economics, Vol. 131 No. 4, pp. 1593-1636.
Barro, R.J. (1991), “Economic growth in a cross section of countries”, Quarterly Journal of Economics,
Vol. 106 No. 2, pp. 407-443.
Bloom, N. (2009), “The impact of uncertainty shocks”, Econometrica, Vol. 77 No. 3, pp. 623-685.
Bloom, N., Floetotto, M., Jaimovich, N., Saporta-Eksten, I. and Terry, S.J. (2018), “Really uncertain
business cycles”, Econometrica, Vol. 86 No. 3, pp. 1031-1065.
Bonaime, A., Gulen, H. and Ion, M. (2018), “Does policy uncertainty affect mergers and acquisitions?”,
Journal of Financial Economics, Vol. 129 No. 3, pp. 531-558.
Caggiano, G., Castelnuovo, E. and Groshenny, N. (2014), “Uncertainty shocks and unemployment
dynamics in US recessions”, Journal of Monetary Economics, Vol. 67 No. C, pp. 78-92.
JES Carmignani, F. (2003), “Political instability, uncertainty and economics”, Journal of Economic Surveys,
Vol. 17 No. 1, pp. 1-54.
49,4
Colak, G., Durnev, A. and Qian, Y. (2017), “Political uncertainty and IPO activity: evidence from US
gubernatorial elections”, Journal of Financial and Quantitative Analysis, Vol. 52 No. 6,
pp. 2523-2564.
Colombo, V. (2013), “Economic policy uncertainty in the US: does it matter for the Euro area?”,
Economics Letters, Vol. 121 No. 1, pp. 39-42.
594
Cragg, J.G. (1983), “More efficient estimation in the presence of heteroscedasticity of unknown form”,
Econometrica, Vol. 51 No. 3, pp. 751-763.
de Mendonça, H.F. (2007), “Towards credibility from inflation targeting: the Brazilian experience”,
Applied Economics, Vol. 39 No. 20, pp. 2599-2615.
de Mendonça, H.F. and Almeida, A.F.G. (2019), “Importance of credibility for business confidence:
evidence from an emerging economy”, Empirical Economics, Vol. 57 No. 6, pp. 1979-1996.
de Mendonça, H.F. and Cacicedo, T. (2015), “Size of government and economic growth in the largest
Latin American country”, Applied Economics Letters, Vol. 22 No. 11, pp. 904-910.
de Mendonça, H.F. and Lima, T.R.V.D.S. (2011), “Macroeconomic determinants of investment under
inflation targeting: empirical evidence from the Brazilian economy”, Latin American Business
Review, Vol. 12 No. 1, pp. 25-38.
Degiannakis, S., Filis, G. and Palaiodimos, G. (2017), “Investments and uncertainty revisited: the case
of the US economy”, Applied Economics, Vol. 49 No. 45, pp. 4521-4529.
Demir, E. and Ersan, O. (2017), “Economic policy uncertainty and cash holdings: evidence from BRIC
countries”, Emerging Markets Review, Vol. 33, pp. 189-200.
Donadelli, M. (2015), “Google search-based metrics, policy-related uncertainty and macroeconomic
conditions”, Applied Economics Letters, Vol. 22 No. 10, pp. 801-807.
Durnev, A. (2010), “The real effects of political uncertainty: elections and investment sensitivity to
stock prices”, available at: https://ssrn.com/abstract51549714.
Gholipour, H.F. (2019), “The effects of economic policy and political uncertainties on economic
activities”, Research in International Business and Finance, Vol. 48, pp. 210-218.
Ghura, D. and Goodwin, B. (2000), “Determinants of private investment: a cross-regional empirical
investigation”, Applied Economics, Vol. 32 No. 14, pp. 1819-1829.
Granger, C.W.J. and Newbold, P. (1974), “Spurious regressions in econometrics”, Journal of
Econometrics, Vol. 2 No. 2, pp. 111-120.
Greenland, A., Ion, M. and Lopresti, J. (2019), “Exports, investment and policy uncertainty”, Canadian
Journal of Economics, Vol. 19, pp. 1248-1288.
Gulen, H. and Ion, M. (2015), “Policy uncertainty and corporate investment”, Review of Financial
Studies, Vol. 29 No. 3, pp. 523-564.
Hall, A.R. (2005), Generalized Method of Moments, 1st ed., Oxford University Press.
Hamilton, J.D. (2018), “Why you should never use the Hodrick-Prescott filter”, Review of Economics
and Statistics, Vol. 100 No. 5, pp. 831-843.
Hendry, D.F. (2001), “Achievements and challenges in econometric methodology”, Journal of
Econometrics, Vol. 100 No. 1, pp. 7-10.
Im, H.J., Park, H. and Zhao, G. (2017), “Uncertainty and the value of cash holdings”, Economics Letters,
Vol. 155, pp. 43-48.
Istiak, K. and Alam, M.R. (2019), “Oil prices, policy uncertainty and asymmetries in inflation
expectations”, Journal of Economics Studies, Vol. 46, pp. 324-334.
Istiak, K. and Alam, M.R. (2020), “US economic policy uncertainty spillover on the stock markets of the
GCC countries”, Journal of Economics Studies, Vol. 47 No. 1, pp. 36-50.
Istiak, K. and Serletis, A. (2018), “Economic policy uncertainty and real output: evidence from the G7 Effects of
countries”, Applied Economics, Vol. 50, pp. 4222-4233.
economic
Jens, C.E. (2017), “Political uncertainty and investment: causal evidence from US gubernatorial
elections”, Journal of Financial Economics, Vol. 124 No. 3, pp. 563-579.
policy
Johnston, J. (1984), Econometric Methods, 3rd ed., McGraw-Hill Book Co, Singapore.
uncertainty
Julio, B. and Yook, Y. (2012), “Political uncertainty and corporate investment cycles”, The Journal of
Finance, Vol. 67 No. 1, pp. 45-83. 595
Julio, B. and Yook, Y. (2016), “Policy uncertainty, irreversibility, and cross-border flows of capital”,
Journal of International Economics, Vol. 103, pp. 13-26.
Kelly, B., P
astor, L. and Veronesi, P. (2016), “The price of political uncertainty: theory and evidence
from the option market”, The Journal of Finance, Vol. 71 No. 5, pp. 2417-2480.
Khan, H. and Upadhayaya, S. (2020), “Does business confidence matter for investment?”, Empirical
Economics, Vol. 59 No. 4, pp. 1633-1665.
Konstantinou, P. and Tagkalakis, A. (2011), “Boosting confidence: is there a role for fiscal policy?”,
Economic Modelling, Vol. 28 No. 4, pp. 1629-1641.
Leduc, S. and Liu, Z. (2016), “Uncertainty shocks are aggregate demand shocks”, Journal of Monetary
Economics, Vol. 82 No. C, pp. 20-35.
Mart
ınez-Serna, M.I. and Navarro, E. (2015), “Interest rate volatility and business cycle expectations”,
International Finance, Vol. 18 No. 1, pp. 69-92.
Montes, G.C. (2013), “Credibility and monetary transmission channels under inflation targeting: an
econometric analysis from a developing country”, Economic Modelling, Vol. 30, pp. 670-684.
Montes, G.C. and Almeida, A.F.G. (2017), “Corruption and business confidence: a panel data analysis”,
Economics Bulletin, Vol. 37 No. 4, pp. 2692-2702.
Montes, G.C. and Bastos, J.C.A. (2013), “Economic policies, macroeconomic environment and
entrepreneurs’ expectations: evidence from Brazil”, Journal of Economics Studies, Vol. 40 No. 3,
pp. 334-354.
Montes, G.C. and Costa, J. (2020), “Effects of fiscal credibility on sovereign risk: evidence using
comprehensive credit rating measures”, International Journal of Emerging Markets, Vol. ahead-
of-print No. ahead-of-print, doi: 10.1108/IJOEM-06-2020-0697.
Montes, G.C. and Souza, I. (2018), “Sovereign default risk, debt uncertainty and fiscal credibility: the
case of Brazil”, The North American Journal of Economics and Finance, Vol. 51 January, p.
100851.
Moore, A. (2017), “Measuring economic uncertainty and its effects”, The Economic Record, Vol. 93
No. 303, pp. 550-575.
Mumtaz, H. and Surico, P. (2018), “Policy uncertainty and aggregate fluctuations”, Journal of Applied
Econometrics, Vol. 33 No. 3, pp. 319-331.
Newey, W. and West, K. (1987), “A simple, positive semi-definite, heteroskedasticity and
autocorrelation consistent covariance matrix”, Econometrica, Vol. 55 No. 3, pp. 703-708.
Nodari, G. (2014), “Financial regulation policy uncertainty and credit spreads in the US”, Journal of
Macroeconomics, Vol. 41, pp. 122-132.
OECD (2013), “Trust in government, policy effectiveness and the governance agenda”, Government at
a Glance 2013, OECD Publishing, Paris. doi: 10.1787/gov_glance-2013-6-en.
P
astor, L. and Veronesi, P. (2012), “Uncertainty about government policy and stock prices”, The
Journal of Finance, Vol. 67 No. 4, pp. 1219-1264.
Phillips, P.C.B. (1986), “Understanding spurious regressions in econometrics”, Journal of Econometrics,
Vol. 33 No. 3, pp. 311-340.
€
Stockhammar, P. and Osterholm, P. (2016), “Effects of US policy uncertainty on Swedish GDP
growth”, Empirical Economics, Vol. 50 No. 2, pp. 443-462.
JES Svensson, J. (1998), “Investment, property rights, political instability: theory and evidence”, European
Economic Review, Vol. 42 No. 7, pp. 1317-1341.
49,4
Tajaddini, R. and Gholipour, H.F. (2020), “Economic policy uncertainty, R&D expenditures and
innovation outputs”, Journal of Economics Studies, Vol. 48 No. 2, pp. 413-427.
Uddin, M.A., Ali, M.H. and Masih, M. (2017), “Political stability and growth: an application of dynamic
GMM and quantile regression”, Economic Modelling, Vol. 64, pp. 610-625.
596 Ventosa-Santaularia, D. (2009), “Spurious regression”, Journal of Probability and Statistics, Vol. 2009,
802975, p. 27, doi: 10.1155/2009/802975.
Windmeijer, F. (2005), “A finite sample correction for the variance of linear efficient two-step GMM
estimators”, Journal of Econometrics, Vol. 126 No. 1, pp. 25-51.
Wooldridge, J.M. (2001), “Applications of generalized method of moments estimation”, The Journal of
Economic Perspectives, Vol. 15 No. 4, pp. 87-100.
Zouhaier, H. and Kefi, M.K. (2012), “Interaction between political instability and investment”, Journal
of Economics and International Finance, Vol. 4 No. 2, pp. 49-54.

Further reading
Hansen, L.P. (1982), “Large sample properties of generalized method of moments estimators”,
Econometrica, Vol. 50 No. 4, pp. 1029-1054.

Corresponding author
Gabriel Caldas Montes can be contacted at: [email protected]
 ADF PP KPSS
Series lag I/T Test 10% Band I/T Test 10% Band I/T Test 1%

ici 1 N 3.094 1.615 2 N 2.269 1.615 10 I 0.698 0.739 Appendix

bci 3 N 2.328 1.615 2 N 2.269 1.615 10 I 0.706 0.739
Gap 3 N 4.130 1.615 5 N 6.355 1.615 9 I 0.040 0.739
EIND_GDP 1 I 2.797 2.576 9 I 2.190 2.576 10 I 0.837 0.739
D(EIND_GDP) 4 N 5.148 1.615 9 I 0.049 0.739
Ipca 1 I 2.207 2.576 7 I 2.033 2.576 10 I 0.252 0.739
D(Ipca) 0 N 6.893 1.615 1 N 6.976 1.615
Eipca 0 I 1.314 2.576 5 I 1.601 2.576 10 I 0.360 0.739
D(Eipca) 0 N 13.351 1.615 6 N 13.447 1.615
Selic 4 I 1.755 2.576 9 I 1.661 2.576 10 I 0.624 0.739
D(Selic) 3 N 3.919 1.615 4 N 3.930 1.615
Exchange 0 I 0.831 2.576 6 I 1.114 2.576 10 I 0.618 0.739
D(Exchange) 0 N 12.329 1.615 6 N 12.517 1.615
Suprim 0 I/T 1.841 3.142 5 I/T 1.985 3.142 10 I/T 0.292 0.216
D(Suprim) 0 N 11.493 1.615 5 N 11.656 2.578 5 I 0.176 0.739
Infr 12 I 2.698 2.576 6 I 3.240 2.576 10 I 0.512 0.739
D(Infr)
CDS 0 I 2.465 2.576 0 I 2.465 2.576 10 I 0.256 0.739
D(CDS) 0 N 11.873 1.615 1 N 11.877 1.615
PU 1 I 0.534 2.576 4 I 0.184 2.576 10 I 0.817 0.739
D(PU) 0 N 7.909 1.615 3 N 7.952 1.615 5 I 0.295 0.739
EPU 0 I/T 6.675 3.143 5 I/T 6.711 3.143 8 I/T 0.285 0.216
D(EPU)_ 22 I 0.074 0.739
INVEST 3 I/T 3.841 3.142 3 I/T 4.019 3.143 9 I/T 0.153 0.216
IRreal 2 I 1.838 2.576 9 I 1.569 2.576 10 I 0.816 0.739
D(IRreal) 1 N 4.595 1.615 7 N 7.969 1.615 9 I 0.072 0.739
EGDP 1 I/T 3.060 3.142 9 I/T 2.172 3.143 10 I/T 0.200 0.216
D(EGDP) 0 N 4.643 1.615 2 N 4.692 1.615
Ibovespa 0 I 1.706 2.576 4 I 1.872 2.576 10 I 0.770 0.739
D(Ibovespa) 0 N 11.119 1.615 4 N 11.140 1.615 4 I 0.083 0.739
Eexchange 3 I 1.509 2.576 8 I 1.174 2.576 10 I 0.558 0.739

(continued )
uncertainty
economic

597
policy
Effects of

Unit root tests

Table A1.
 JES
49,4

598

Table A1.
ADF PP KPSS
Series lag I/T Test 10% Band I/T Test 10% Band I/T Test 1%

D(Eexchange) 2 N 4.395 1.615 2 N 5.325 1.615

Credit 3 I 2.146 2.576 8 I 2.439 2.576 10 I 1.507 0.739
D(Credit) 2 I/T 4.995 3.143 5 I/T 9.924 3.143 7 I/T 0.231 0.216
Debt 6 I 0.245 2.576 5 I 0.872 2.576 10 I 0.375 0.739
D(Debt) 5 I/T 3.833 3.143 0 I/T 17.225 3.143
Note(s): ADF–the final choice of lag was made based on Schwarz information criterion. PP and KPSS tests–Band is the bandwidth truncation chosen for the
Bartlett kernel. “I” denotes intercept; “I/T” denotes intercept and trend and; “N” denotes none
Variables Mean Median Max Min SD
Effects of
economic
bci 0.164 0.003 1.226 2.358 1.006 policy
ici 0.187 0.004 1.323 2.633 1.006
bci* 0.165 0.001 1.226 2.358 1.000 uncertainty
ici* 0.189 0.000 1.323 2.633 1.000
PU 0.001 0.507 4.112 2.011 1.730
EPU 163.046 136.492 676.955 22.296 102.383 599
Exchange 2.361 2.202 4.043 1.556 0.622
CDS 219.973 171.234 805.419 62.157 137.492
Credit 41.170 43.745 53.860 23.980 9.496
Debt 64.537 63.000 81.550 57.030 5.932
Eexchange 2.454 2.268 4.323 1.623 0.682
Eipca 5.355 5.383 7.557 3.095 0.997
EGDP 2.625 3.389 5.910 2.990 1.997
EIND_GDP 2.837 3.365 8.834 3.737 2.715
Gap 0.068 0.093 6.036 7.313 2.548
ICM 0.559 0.555 0.986 0.000 0.298
Infr 14.263 14.228 15.886 11.829 0.948
INVEST 3.031 3.765 30.730 19.550 10.391
Ipca 5.946 5.905 10.710 2.460 1.750
IRreal 6.968 6.641 14.838 1.374 3.190
Selic 12.318 11.780 19.750 7.110 3.146
Suprim 1.797 2.545 4.080 3.050 2.081 Table A2.
Note(s): bci* and ici* are standardized series of BCI and ICI (mean zero and standard deviation 1), respectively Descriptive statistics
JES Table Equation Instruments
49,4
Table 2 Eq. (1) - GMM ICI (1 to 6) GAP (1 to 5) E_IND_GDP (5) D(IPCA (1)) D(IPCA(2))
D(IPCA(3)) D(E_IPCA(2))
D (E_IPCA (3)) D(E_IPCA (4)) D(E_IPCA(5)) D(E_IPCA(6))
D(SELIC(3)) D (EXCHANGE(1))
D (EXCHANGE (2)) D (SUPRIM (2)) D (SUPRIM (3)) D(SUPRIM (4))
600 D(SUPRIM(5))
D (SUPRIM (6)) INFR (1 to8) D(CDS (4)) D(PU(1))
Eq. (2) - GMM ICI (1 to 6) GAP (1 to 5) E_IND_GDP (5) D(IPCA (1)) D(IPCA(2))
D(IPCA(3)) D(E_IPCA(2))
D(E_IPCA (3)) D(E_IPCA(4)) D(E_IPCA(5)) D(E_IPCA(6))
D(SELIC(3)) D(EXCHANGE(1))
D(EXCHANGE(2)) D(SUPRIM(2)) D(SUPRIM(3)) D(SUPRIM(4))
D(SUPRIM(5))
D(SUPRIM(6)) INFR(1 to8) D(CDS(4)) EPU(1 to 2)
Eq. (3) - GMM ICI (1 to 5) GAP(1 to 5) E_IND_GDP(5 to 7) D(IPCA(1))
D(IPCA(2)) D(IPCA(3)) D(E_IPCA(2))
D(E_IPCA(3)) D(SELIC(3)) D(SELIC(4)) D(EXCHANGE(1))
D(SUPRIM(2)) INFR(1 to8) D(CDS(4))
D(CDS(5)) D(CDS(6)) D(PU(1)) D(PU(2)) D(PU(3)) D(PU(4))
EPU(1)
Eq. (4) - GMM ICI (1 to 5) GAP(1 to 5) E_IND_GDP(5 to 7) D(IPCA(1))
D(IPCA(2)) D(IPCA(3)) D(E_IPCA(2))
D(E_IPCA(3)) D(SELIC(3)) D(SELIC(4)) D(EXCHANGE(1))
D(SUPRIM(2)) INFR(1 to8) D(CDS(4))
D(CDS(5)) D(CDS(6)) D(ICM(2)) D(PU(1)) D(PU(2)) D(PU(3))
D(PU(4)) EPU(1)
Eq. (1) - GMM-2 ICI (1 to 6) GAP(1 to 5) E_IND_GDP(5 to 11) D(IPCA(1))
D(IPCA(2)) D(IPCA(3)) D(E_IPCA(2))
D(SELIC(3)) D(EXCHANGE(1)) D(SUPRIM(2)) D(SUPRIM(3))
D(SUPRIM(4)) D(SUPRIM(5))
D(SUPRIM(6)) D(SUPRIM(7)) INFR(1 to9) D(CDS(4)) D(PU(1))
D(PU(2))
Eq. (2) - GMM-2 ICI (1 to 4) GAP(1 to 8) E_IND_GDP(5 to 11) D(IPCA(1))
D(E_IPCA(2)) D(E_IPCA(3))
D(E_IPCA(4)) D(E_IPCA(5)) D(SELIC(3)) D(SELIC(4))
D(EXCHANGE(1)) D(EXCHANGE(2))
D(EXCHANGE(3)) D(SUPRIM(2)) INFR(1 to8) D(CDS(4))
D(CDS(5)) EPU(1 to 2)
Eq. (3) - GMM2 ICI (1 to 4) GAP(1 to 7) E_IND_GDP(5 to 11) D(IPCA(1))
D(E_IPCA(2)) D(E_IPCA(3))
D(E_IPCA(4)) D(E_IPCA(5)) D(E_IPCA(6)) D(SELIC(3))
D(EXCHANGE(1)) D(EXCHANGE(2))
D(EXCHANGE(3)) D(SUPRIM(2)) D(SUPRIM(3)) INFR(1 to10)
D(CDS(4)) D(PU(1)) EPU(1 to 3)
Eq. (4) - GMM2 ICI (1 to 3) GAP(1 to 7) E_IND_GDP(5 to 9) D(IPCA(1))
D(IPCA(2)) D(E_IPCA(2)) D(E_IPCA(3))
D(SELIC(3)) D(EXCHANGE(1)) D(EXCHANGE(2))
D(EXCHANGE(3)) D(EXCHANGE(4)) D(SUPRIM(2))
INFR(1 TO13) D(CDS(4)) D(ICM(2)) D(ICM(3)) D(PU(1)) D(PU(2))
Table A3. D(PU(3)) D(PU(4)) EPU(1)
List of instruments
(GMM and GMM-2) (continued )
Table Equation Instruments
Effects of
economic
Table 3 Eq. (5) - GMM BCI (1 to 6) GAP(1 to 5) E_IND_GDP(5) D(IPCA(1)) D(IPCA(2)) policy
D(IPCA(3)) D(E_IPCA(2))
D(E_IPCA(3)) D(E_IPCA(4)) D(E_IPCA(5)) D(E_IPCA(6)) uncertainty
D(SELIC(3)) D(EXCHANGE(1))
D(EXCHANGE(2)) D(SUPRIM(2)) D(SUPRIM(3)) D(SUPRIM(4))
D(SUPRIM(5)) 601
D(SUPRIM(6)) INFR(1 to8) D(CDS(4)) D(PU(1))
Eq. (6) - GMM BCI (1 to 6) GAP(1 to 5) E_IND_GDP(5) D(IPCA(1)) D(IPCA(2))
D(IPCA(3)) D(E_IPCA(2))
D(E_IPCA(3)) D(E_IPCA(4)) D(E_IPCA(5)) D(E_IPCA(6))
D(SELIC(3)) D(EXCHANGE(1))
D(EXCHANGE(2)) D(SUPRIM(2)) D(SUPRIM(3)) D(SUPRIM(4))
D(SUPRIM(5))
D(SUPRIM(6)) INFR(1 to8) D(CDS(4)) EPU(1 to 2)
Eq. (7) - GMM BCI (1 to 5) GAP(1 to 5) E_IND_GDP(5 to 7) D(IPCA(1))
D(IPCA(2)) D(IPCA(3)) D(E_IPCA(2))
D(E_IPCA(3)) D(SELIC(3)) D(SELIC(4)) D(EXCHANGE(1))
D(SUPRIM(2)) INFR(1 to8) D(CDS(4))
D(CDS(5)) D(CDS(6)) D(PU(1)) D(PU(2)) D(PU(3)) D(PU(4))
EPU(1)
Eq. (8) - GMM BCI (1 to 5) GAP(1 to 5) E_IND_GDP(5 to 7) D(IPCA(1))
D(IPCA(2)) D(IPCA(3)) D(E_IPCA(2))
D(E_IPCA(3)) D(SELIC(3)) D(SELIC(4)) D(EXCHANGE(1))
D(SUPRIM(2)) INFR(1 to8) D(CDS(4))
D(CDS(5)) D(CDS(6)) D(ICM(2)) D(PU(1)) D(PU(2)) D(PU(3))
D(PU(4)) EPU(1)
Eq. (5) - GMM2 BCI (1 to 6) GAP(1 to 5) E_IND_GDP(5 to 11) D(IPCA(1))
D(IPCA(2)) D(IPCA(3)) D(IPCA(4))
D(E_IPCA(2)) D(E_IPCA(3)) D(SELIC(3)) D(EXCHANGE(1))
D(EXCHANGE(2)) D(SUPRIM(2))
D(SUPRIM(3)) D(SUPRIM(4)) D(SUPRIM(5)) D(SUPRIM(6))
D(SUPRIM(7)) INFR(1 to8) D(CDS(4))
D(CDS(5)) D(PU(1)) D(PU(2))
Eq. (6) - GMM2 BCI (1 to 6) GAP(1 to 5) E_IND_GDP(5 to 11) D(IPCA(1))
D(IPCA(2)) D(IPCA(3)) D(E_IPCA(2))
D(SELIC(3)) D(EXCHANGE(1)) D(EXCHANGE(2)) D(SUPRIM(2))
D(SUPRIM(3)) D(SUPRIM(4))
D(SUPRIM(5)) D(SUPRIM(6)) D(SUPRIM(7)) INFR(1 to9)
D(CDS(4)) EPU(1 to 2)
Eq. (7) - MM2 BCI (1 to 6) GAP(1 to 7) E_IND_GDP(5 to 10) D(IPCA(1))
D(E_IPCA(2)) D(E_IPCA(3))
D(E_IPCA(4)) D(E_IPCA(5)) D(SELIC(3)) D(EXCHANGE(1))
D(EXCHANGE(2)) D(EXCHANGE(3))
D(SUPRIM(2)) D(SUPRIM(3)) D(SUPRIM(4)) INFR(1 to10)
D(CDS(4)) D(PU(1)) D(PU(2)) EPU(1 to 2)
Eq. (8) - GMM2 BCI (1 to 3) GAP(1 to 7) E_IND_GDP(5 to 9) D(IPCA(1))
D(IPCA(2)) D(E_IPCA(2)) D(E_IPCA(3))
D(SELIC(3)) D(EXCHANGE(1)) D(EXCHANGE(2))
D(EXCHANGE(3)) D(EXCHANGE(4)) D(SUPRIM(2))
INFR(1 to12) D(CDS(4)) D(ICM(2)) D(ICM(3)) D(PU(1)) D(PU(2))
D(PU(3)) D(PU(4)) EPU(1)

(continued ) Table A3.

JES Table Equation Instruments
49,4
Table 4 Eq. (9) - GMM INVEST(1 to 6) D(IR_REAL(5)) D(IR_REAL(6)) D(IR_REAL(7))
D(IR_REAL(8)) D(IR_REAL(9))
D(IR_REAL(10)) GAP(1 to 4) D(E_GDP(5)) D(E_IPCA(5))
D(E_IPCA(6)) D(E_IPCA(7)) D(E_IPCA(8))
D(IPCA(4)) D(IPCA(5)) D(IPCA(6)) D(IPCA(7)) D(IPCA(8))
602 D(IPCA(9)) D(IBOVESPA(5)) D(IBOVESPA(6))
D(IBOVESPA(7)) D(IBOVESPA(8)) D(EXCHANGE(5))
D(EXCHANGE(6)) D(E_EXCHANGE(5))
D((E_EXCHANGE (6)) D(E_EXCHANGE(7)) D(E_EXCHANGE(8))
D(E_EXCHANGE(9))
D(E_EXCHANGE(10)) D(CREDIT(7)) D(CREDIT(8)) D(DEBT(4))
ICI(4 to 5)
Eq (10) - GMM INVEST(1 to 4) D(IR_REAL(5)) D(IR_REAL(6)) D(IR_REAL(7))
D(IR_REAL(8)) GAP(1 to 7)
D(E_GDP(5)) D(E_GDP(6)) D(E_IPCA(5)) D(E_IPCA(6))
D(E_IPCA(7)) D(E_IPCA(8)) D(IPCA(4))
D(IPCA(5)) D(IPCA(6)) D(IPCA(7)) D(IBOVESPA(5))
D(IBOVESPA(6)) D(IBOVESPA(7)) D(EXCHANGE(5))
D(EXCHANGE(6)) D(E_EXCHANGE(5)) D((E_EXCHANGE (6))
D(E_EXCHANGE(7)) D(CREDIT(7))
D(CREDIT(8)) D(DEBT(4)) D(DEBT(5)) BCI(4 to 7)
Eq. (9) - GMM2 INVEST(1 to 8) D(IR_REAL(5)) GAP(1 to 8) D(E_GDP(5))
D(E_IPCA(5)) D(E_IPCA(6))
D(E_IPCA(7)) D(IPCA(4)) D(IPCA(5)) D(IPCA(6)) D(IPCA(7))
D(IPCA(8)) D(IPCA(9))
D(IPCA(10)) D(IBOVESPA(5)) D(IBOVESPA(6)) D(IBOVESPA(7))
D(IBOVESPA(8)) D(IBOVESPA(9))
D(EXCHANGE(5)) D(E_EXCHANGE(5)) D(CREDIT(7))
D(CREDIT(8)) D(DEBT(4)) D(DEBT(5))
D(DEBT(6)) D(DEBT(7)) ICI(4 to 5) D(SELIC(1)) D(SUPRIM(1))
Eq (10) - GMM-2 INVEST(1 to 8) D(IR_REAL(5)) GAP(1 to 8) D(E_GDP(5))
D(E_IPCA(5)) D(E_IPCA(6)) D(E_IPCA(7))
D(IPCA(4)) D(IPCA(5)) D(IPCA(6)) D(IPCA(7)) D(IPCA(8))
D(IPCA(9)) D(IPCA(10))
D(IBOVESPA(5)) D(IBOVESPA(6)) D(IBOVESPA(7))
D(IBOVESPA(8)) D(IBOVESPA(9))
D(EXCHANGE(5)) D(E_EXCHANGE(5)) D(CREDIT(7))
D(CREDIT(8)) D(DEBT(4)) D(DEBT(5))
Table A3. D(DEBT(6)) D(DEBT(7)) BCI(4 to 5) D(SELIC(1))