Corporate governance, pressure-insensitive institutional investors and firm

performance: Evidence from India

Abstract
The study analyses the role of corporate governance mechanisms, audit

committee structure and institutional investors in enhancing firm performance. Unlike in

developed economies where firm ownership is widely dispersed, firms in emerging

economies such as India have substantial promoter shareholdings (often in majority or

close to majority). Given the promoter control of Indian companies, the role of

institutional investors as external monitors is analysed. Following Brickely et al (1988),

the study categorises institutional investors as pressure-sensitive and pressure-insensitive

institutional investors. The study finds that increased ownership of pressure-insensitive

institutional investors is positively associated with firm performance. At the same time,

increased ownership of pressure-sensitive institutional investors is negatively associated

with firm performance. These results are consistent with the view that pressure-

insensitive institutional investors are more effective monitors when compared to

pressure-sensitive institutional investors. The evidence also suggests that increased audit

committee meetings and audit committee independence is associated with enhanced firm

performance.

Keywords: Corporate governance, Institutional investors, Audit committee, Firm

ownership, Firm performance, India

 Corporate governance, pressure-insensitive institutional investors and firm
performance: Evidence from India

1. Introduction
Corporate governance mechanisms can be viewed as internal and external governance

mechanisms. For instance, board of directors and compensation structures can be viewed as

part of internal governance mechanisms. Other mechanisms such as market for corporate

control, the competitive environment and legal environment can be perceived as part of

external governance mechanisms (Walsh and Seward, 1990). Proposed corporate governance

reforms such Sarbanes-Oxley Act (2002) in US or Companies Act (2013) in an emerging

economy such as India emphasize the need to strengthen monitoring of management

decisions. One of the key objectives of these proposed corporate governance reforms is to

include more “independent” actors on the board. At the same time, in developing economies

such as India, the evidence with regards to board independence and firm performance is

counter intuitive. For example, recent studies (Bansal & Sharma, 2016; Arora and Sharma,

2016; Vinjamury (2020)) report a negative relationship between board independence and firm

performance. A one-size-fits-all approach may not be appropriate in terms of corporate

governance mechanisms.

In addition to board of directors, institutional investors can be viewed as monitors that

work in the best interest of the shareholders. In economies such as India, where firm

shareholdings are dominated by promoters (for example see, Jameson et al., 2014) the role of

institutional investors needs closer examination. Prior studies (see Brickley et al (1988) and

Almazan (2005)) posit that all institutional investors are not equal. For example, institutional

investors such as banks and insurance companies may have existing relationships with firms.

In order to continue their business relationship, these institutional investors may not be

willing to confront and challenge management decisions. These institutional investors can be
considered as pressure-sensitive institutional investors. On the other hand, institutional

investors such as mutual funds and foreign institutional investors are less likely to have

business relationships and are in a better position to monitor and enforce discipline on

corporate managers. These institutional investors can be considered as pressure- insensitive

institutional investors.

In this backdrop, the study seeks to analyse the role of institutional investors in

enhancing firm performance. Also, the study seeks to analyse whether all institutional

investors are created equal in the context of an emerging economy such as India. Specifically,

the study seeks to analyse the relationship between pressure-sensitive institutional investors

and pressure-insensitive institutional investors with respect to firm performance.

The rest of the paper is structured as follows. Section 2 provides a review of the

literature. Section 3 discusses the objectives of the study and lists the control variables which

may potentially influence firm performance. Section 4 discusses data and methodology.

Section 5 presents empirical results and Section 6 concludes the study.

2. Literature review

Prior studies have viewed institutional investors in the role as corporate monitors. In

this context, Grossman and Hart (1980) posit that large shareholders such as institutional

investors have incentives to monitor as they can achieve sufficient benefits by monitoring. In

a similar vein, Shlifer and Vishny (1986) argue that larger shareholders have a greater

incentive to monitor the management than the board of directors since board of directors may

have little or no wealth invested in the firm. Consistent with this view, other studies such as

Nesbitt (1994), McConnell and Servaes (1990), Smith (1996) show that monitoring by

institutional investors can stem self-serving behaviour of the managers and can enhance firm

performance.
 It can be argued that institutional investors with larger shareholdings will have greater

incentive to monitor the managers. In this context, Maug (1998) highlights that monitoring by

institutional investors is partly a function of the size of their shareholdings. When

institutional investors hold larger shareholdings which may be potentially illiquid, they have

a greater incentive to monitor. On the other hand, if the institutional investors hold fewer

shares, they can liquidate their holding quickly and may not have a strong incentive to

monitor. Consistent with the later view, studies such as Coffee (1991), Bhide (1994) and

Maug (1998) document that institutional investors are likely to be driven by short-term

profits and may have less incentive to monitor the management.

Never the less, with greater emphasis given to corporate governance mechanisms, the

role of institutional investors has received greater attention in the developed economies.

Studies have highlighted the role of institutional investors in “taming” the management. For

example, Parrino et al. (2003) show that institutional selling is linked to forced CEO turnover

and that the replacement is likely to be an outsider. Similarly, Chung et al. (2002) document

that large institutional shareholdings in a firm may hinder managers from adopting

opportunistic discretionary accrual choices. Other studies have analysed the role of

institutional investors on firm performance. Prior studies such as Karpoff et al. (1996),

Duggal and Millar (1999) Agrawal and Knoeber (1996) and Faccio and Lasfer (2000) do not

find a significant relationship between institutional investor ownership and firm performance.

However, McConnell and Servaes (1990) is one of the early studies that document a positive

relationship between institutional ownership and Tobin’s Q used as a proxy for firm

performance. In a similar vein, Nesbitt (1994), Smith (1996) and Del Guercio and Hawkins

(1999) find a positive association between institutional investor ownership and different

measures of firm performance.

 Brickley et al (1988) provide a framework where they categorize institutional

investors as pressure-sensitive or pressure-insensitive. They argue that pressure-insensitive

institutional investors are more likely to monitor and discipline the managers. Pressure-

sensitive institutional investors who are likely to have existing business relationships with the

firm are less likely to be effective monitors. Consistent with this view, Almazan et al. (2005)

shows that institutional ownership and executive compensation are negatively related and the

relationship is stronger for pressure-sensitive institutional investors. In a similar vein, Cornett

et al. (2007) document a positive relationship between pressure-insensitive institutional

investors and corporate operating performance. However, the presence of pressure-sensitive

institutional investors does not seem to have an impact on firm’s operating cash flow returns.

3. Objectives of the study and control variables

Many studies exploring the relationship between pressure-sensitive/ pressure-

insensitive institutional investors have focused their attention on large firms in developed

economies such as US where the firms are characterised by dispersed share ownership and

are not dominated by promoter shareholdings. On the other hand, most Indian firms have

concentrated ownership with substantial promoter shareholdings (for example see, Jameson

et al (2014)). Given the difference in the ownership structure, the study intends to explore the

relationship between institutional ownership and firm performance in the Indian context.

Specifically, the objectives of the study are as follows:

1. To understand the relationship between institutional ownership and firm performance.

2. To understand the relationship between pressure-sensitive institutional investors and

firm performance.

3. To understand the relationship between pressure-insensitive institutional investors and

firm performance.
3.1 Control variables

The primary objective of this study is to analyse the relationship between institutional

ownership and firm performance. The following variables will be used as control variables in

the study

3.1.1 Board size

In terms of the board size, previous studies in developed economies have documented

that large boards negatively impact the value of the firm (For example, see Yermack, 1996).

In a similar vein, Eisenberg et al. (1998) find a negative and significant relationship between

board size and profitability in a sample of small and midsized Finnish firms. From theoretical

perspective, Jensen (1993) argues that for larger corporate boards, the problems of

communication and coordination increase with increase in number of board members.

However, an argument can be made that a bigger board would bring more knowledge and

expertise to the table which would be value enhancing for the shareholders (For example, see

De Oliveira et al, 2012; Saibaba and Ansari ,2002). Put differently, smaller boards may not

be well equipped for making strategic changes when considering alternatives for firm growth

(Hambrick et al., 2008). The latter view appears to be more relevant in the Indian context.

Studies on Indian firms have shown a significantly positive relationship between larger

boards and measures of firm performance (for example, see Arora and Sharma (2016); Bansal

and Sharma (2016); Vinjamury (2020)).

3.1.2 Board Independence

 Conventional wisdom may suggest that greater board independence may improve

monitoring and enhance decision making process. For example, Weisbach (1988) documents

that boards which are more independent than those that are insider dominated are

significantly more likely to remove a CEO based on CEO’s poor performance. This study

also shows that greater board independence enhances firm value through these CEO changes.

In a similar vein, Byrd and Hickman (1992) study the relationship between characteristics of

board members of bidding firms and the associated shareholder wealth effects in tender offer

bids. The study shows that lower negative returns to shareholders are found with respect to

firms that have boards constituted in such a way that at least half the board members are

independent or unaffiliated. This evidence, the authors argue may be consistent with the view

that greater board independence enhances shareholders’ wealth. In another study, Fama and

Jensen (1983) posit that independent outside directors on a firm’s board may have an

incentive to establish a reputation as experts in decision making. From this perspective,

independent directors may seek to enhance their attractiveness and credibility as candidates

for subsequent board appointments at other firms by maintaining a favourable reputation as

active external monitors. However, other studies (Koerniandi & Tourani, (2012); Leung et

al., (2014); Shan & McIver, (2011)) posit that greater board independence need not

necessarily be positively associated with firm performance. The argument in this context is

that independent directors may not have enough information and knowledge about the firm

and may not act as effective monitors.

3.1.3 CEO duality

Prior studies have shown that corporate boards where the roles of CEO and chairmanship are

helmed by two different individuals, the separation of roles is associated with enhanced firm

operating performance (Bhagat and Bolton (2002)). At the same time, other studies (Daily

and
Dalton, (1997) find no significant difference between firm performance irrespective of any

such separation in roles at the top.

3.1.4 Audit committee independence

Auditing plays an important role in strengthening corporate governance mechanisms.

In its function, audit committee is expected to monitor aspects pertaining to accounting,

auditing of financial statements and subsequent reporting. Therefore, an argument can be

made that greater audit committee independence is essential for its effective performance

(Cohen, 2011). Empirically, Beasley, 1996 documents that firms greater audit committee

independence are less likely to be victims of financial frauds. Greater audit committee

independence also allows earnings management can be curtailed (Bukit and Iskandar,

(2009)). In a similar vein, Abbott (2002) documents an inverse relationship between greater

audit committee independence and earnings management.

3.1.5 Audit committee meetings

In addition to audit committee independence, frequency of audit committee meetings

may help enhance monitoring. For example, Menon and Williams (1994) show that greater

audit committee independence and frequency of audit committee meetings led to better

monitoring of the firm. This enhanced monitoring can could enhance the performance of the

firm. In a similar vein, DeZoort et al. (2002) document that more frequent audit committee

meetings allow the audit committees to be more careful in safeguarding the interest of its

investors. More recently, Al-Mamun et al. (2014) posit that regular audit committee meetings

could reduce information asymmetry and agency problems by providing timely information

to the investors.

3.1.6 Leverage
 Many studies have documented that financial leverage is a significant determinant for

firm performance. From a theoretical perspective, if agency costs of the firm can be reduced

due to inclusion of debt in the capital structure, we should expect to see a positive association

between financial leverage and performance (Jensen, 1986). However, studies in developing

economies such as India document a negative and significant relationship between financial

leverage and measures of firm performance (Bansal and Sharma, 2016; Arora and Sharma,

2016 etc.).

3.1.7 Firm Age

Firm age may also be associated with measures of firm performance. In this regard,

many studies on Indian firms document a negative association between firm age and firm

performance measured using Tobin’s Q (for example see, Arora and Bansal (2016)).

4. Data and Methodology

The data for the study was collected from CMIE (Centre for Monitoring Indian

Economy) Database. To undertake the analysis, data was collected for non-financial firms

included in NSE (National Stock Exchange) 500 in India for the time period 2008 to 2017.

To determine the impact of institutional investors with respect to measures of firm

performance, two models are analysed in this study. In the first model, the influence of

institutional investors (without categorising institutional investors as pressure-sensitive and

pressure-insensitive) on firm performance is analysed. In the alternative model, the influence

of pressure-sensitive institutional investors and pressure-insensitive institutional investors on

firm performance is analysed.

 The data for the analysis comprises of unbalanced panel data. Panel data models were

employed to examine fixed and/or random effects. Since panel data includes repeated

observation over time, fixed/random models allow us to examine the relevant changes, if at

all any, with respect to firm over time. Fixed effect regression models was employed for the

analysis in the study. Hausman (1978) test was used to validate the use of fixed effects model

and reject a random effects model. To avoid the impact of outliers in the study, independent

variables were winsorized at 1% level.

The variables used for models are: board size (BSIZE), board independence (BIND),

audit committee independence (AIND), audit committee meetings (AMEET) and institutional

investor ownership (INT_INV). Four performance indicators were analyzed as part of the

study. Return on Assets (ROA), Return on Equity (ROE) and Net Profit Margin (NPM) are

considered as part of accounting performance measures. In terms of a market performance

measure, adjusted Tobin’s Q (TQ) was used. TQ was obtained using similar the methodology

adopted in Gompers et al. (2003). Leverage (LEV), firm age (FAGE), and firm size

(LOG_TA) are used as control variables. Table I provides a detailed description of these

variables.

To test the effects of institutional ownership on firm performance, the following

models are considered:

Y it = α 0 + βX it + II it +Υ C it +ε it (1)

Y it = α 0 + βX it + PSPI it +Υ C it + ε it (2)

In the models above, Y it is the dependent variable and represents firm performance

measure, X it is the set of corporate governance variables used in the study C it is the set of

control variables for firm i at time t. . II it is institutional investor ownership for firm i at time t.
Initially, the relationship between total institutional ownership and firm performance is

analyzed. For more detailed analysis, in model (2) the relationship between pressure-sensitive

institutional investors and pressure-insensitive institutional investors on firm performance is

analyzed. PS PIit is the set of pressure-sensitive and pressure insensitive institutional ownership

for firm i at time t. α 0 is the intercept term to be estimated. β , and Υ represent the parameters

to be estimated for corporate governance, institutional ownership and control variables

respectively. ε it is the error term in the model.

[Insert Table I. here]

5. Empirical results and analysis

Table II documents summary statistics of the variables used in the study. The median

board size (BSIZE) of the firms used in the analysis is 11. The median board independence

(BIND) is 50%. The mean and median non-promoter institutional investor ownership

(INT_INV) is 22.28 % and 20.82% respectively. Of the total institutional ownership, the

mean pressure sensitive (P_SENSITIVE) and pressure insensitive (P_INSENSITIVE)

institutional ownership represents 5.25% and 16.85% respectively. The median audit

committee independence (AIND) for the sample firms is 80%.

[Insert Table II. here]

Multi-collinearity may be a potential concern with regard to the variables used in the

study. Therefore, correlation analysis for variables used in the study was carried out. Table III

documents the results of Pearson correlation coefficients. From this analysis, multi-

collinearity does not appear to be a concern in the study. The highest value is 0.406 between

natural log of total assets (LOG_TA) and non-promoter pressure-sensitive institutional

shareholdings (P_SENSITIVE). As a further check, Variance Inflation Factor (VIF) analysis

was done. None of the VIF values were greater than 1.547 and were within the acceptable

limits. Given this background, multi-collinearity is not a concern for the study.

Tables IV documents results of fixed effects panel data regressions. As mentioned

before, performance measures (TQ, ROA, ROE and NPM) are used for the analysis.

Empirical results show a significant positive association between board size and Tobin’s Q

(TQ). The remaining accounting performance measures NPM, ROE and ROA are not are not

significantly related to board size. This result is consistent with other recent studies in the

Indian context (see, Bansal and Sharma, 2016).

CEO duality (CEO_DUAL) is positively associated with both ROA and ROE.

However, CEO duality is not significantly associated with TQ and NPM. In terms of audit

committee characteristics, number of audit committee meeting (AMEET) is positively

associated with TQ and audit independence (AIND) is positively associated with NPM. Non-

promoter institutional ownership (INT_INV) is positively associated with three measures TQ,

ROA and NPM used in the study suggesting an overall positive impact of the institutional

ownership on firm performance.

[Insert Table IV. here]

The results of fixed effects panel data regression using pressure-sensitive and

pressure-insensitive institutional investors are reported in Tables V. The relationship between

corporate governance variables and performance is consistent with the results documented in

Table IV. However, now board independence (BIND) is negatively associated with TQ. With

respect to institutional ownership variables, pressure-insensitive institutional investors

(P_INSENSITIVE) are positively associated with measures of firm performance TQ, ROA

and NPM. On the other hand, increased ownership of pressure-sensitive institutional

investors is negatively associated with measures of firm performance (TQ, ROA and ROE).
 [Insert Table V. here]

6. Conclusions

Prior studies posit that all institutional investors are not equal. Some institutional

investors, in order to continue their existing business relationship with the firm, may not be

willing to confront and challenge management decisions. These institutional investors can be

considered as pressure-sensitive institutional investors. On the other hand, institutional

investors who are less likely to have business relationships with the firms are in a better

position to monitor and enforce discipline on corporate managers. The latter can be

considered as pressure-insensitive institutional investors.

Given this background, the study analysed the role of institutional investors in

monitoring and subsequently enhancing firm performance. Specifically, the study analysed

whether all institutional investors are created equal in the context of an emerging economy

such as India. Unlike in developed economies such as US where ownership is widely

dispersed, firms in the emerging economy such as India have substantial promoter

shareholdings (often in majority or close to majority). In this context analysing the role of

institutional investors as effective monitors becomes important. The results of the study show

that increased ownership of pressure-insensitive institutional investors is positively associated

with firm performance. At the same time, increased ownership of pressure-sensitive

institutional investors is negatively associated with firm performance. These results are

consistent with the view that pressure-insensitive institutional investors are more effective

monitors when compared to pressure-sensitive institutional investors. In addition, the study

provides some support to the notion that increased audit committee meetings and audit

committee independence is associated with enhanced firm performance.

Table I. Description of the variables used in the study
Variable Description How is it measured
Firm performance variables
(Total assets + market capitalization - book value of equity -
TQ Adjusted Tobin's q
deferred tax liability) as a proportion of total assets
ROA Return on assets Net Income as a proportion of total assets
Net Income as a proportion of (Paid-up equity capital + reserves
ROE Return on equity
and funds)
NPM Net profit margin Net Income as a proportion of Sales

Corporate governance variables

BSIZE Board size Total number of directors on the company’s board
Number of independent directors on board as a proportion of
BIND Board independence
number of directors on company’s board
BODMEET Board meetings Frequency of annual board meetings
Dummy variable. Variable equals 1 when CEO is also the board
CEO_DUAL Chief executive officer dual role
chairman; 0 otherwise
AIND Audit committee independence Percentage of independent directors on the audit committee
Number of audit committee members times no of audit
AMEET Audit committee meetings
committee meetings in a financial year

Institutional Ownership variables

INT_INV Institutional investor ownership Percentage of shares held by non-promoter institutions
Pressure insensitive institutional Percentage of shares held by non-promoter mutual funds and
P_INSENSITIVE
investors foreign institutional investors
Pressure sensitive institutional Percentage of shares held by non-promoter banks and financial
P_SENSITIVE
investors institutions

Panel D: Control variables

FAGE Firm age Age of the firm (in years) since its incorporation
LEV Financial leverage Borrowings as a proportion of total assets
LOG_TA Logarithm of firm’s total assets Natural logarithm of firm’s total assets
Note: The table describes the variables used in the study
Table II. Descriptive statistics

Variable Mean Median Std Dev

BSIZE 11.113 11.000 3.436
BIND 0.501 0.500 0.138
CEO_DUAL 0.011 0.000 0.103
AMEET 16.987 15.000 7.707
AIND 0.793 0.800 0.219
LEV 0.215 0.197 0.186
LOG_TA 10.497 10.384 1.406
FAGE 41.831 34.000 24.327
INT_INV 22.282 20.825 13.528
P_INSENSITIV
16.854 15.765 11.857
E
P_SENSITIVE 5.258 3.115 6.444
Note: This table reports the mean (Mean), median (Median) and standard deviation (Std.
Dev) of the variables used for the analysis.
Table III. Correlation matrix

  BSIZE BIND CEO_DUAL AMEET AIND LEV LOG_TA FAGE P_INSENSITIVE P_SENSITIVE

BSIZE 1.000

BIND -0.197 1.000

CEO_DUAL -0.045 0.067 1.000

AMEET 0.274 -0.051 -0.018 1.000

AIND -0.012 0.431 0.027 0.058 1.000

LEV 0.026 0.089 -0.015 -0.051 0.083 1.000

LOG_TA 0.430 -0.065 -0.051 0.272 -0.007 0.132 1.000

FAGE 0.066 0.007 -0.041 0.130 0.032 -0.057 0.122 1.000

P_INSENSITIVE 0.103 0.124 -0.011 0.050 0.099 -0.162 0.203 -0.029 1.000

P_SENSITIVE 0.250 -0.133 -0.022 0.203 -0.035 -0.015 0.406 0.316 0.002 1.000

Note: Pearson correlation coefficients among independent variables used for the analysis are reported
above.
Table IV. Regression analysis using fixed effects model

Independent Model 1 Model 2 Model 3 Model 4

Variables (TQ) (ROA) (ROE) (NPM)
Intercept -3.82962*** 0.615014*** 1.231764*** 51.1768***
( -6.32 ) 17.73 ( 6.42 ) ( 5.08 )
BSIZE 0.032231** -0.00071 0.002855 -0.23958
( 2.5 ) ( -0.96 ) ( 0.7 ) ( -1.12 )
BIND -0.37357 -0.02489 -0.08132 -4.91989
( -1.36 ) ( -1.59 ) ( -0.94 ) ( -1.08 )
CEO_DUAL 0.298961 0.030739* 0.170737* 0.759452
( 1.01 ) ( 1.82 ) ( 1.82 ) ( 0.15 )
AMEET 0.008573** -0.00008 0.000953 0.015114
( 2.09 ) ( -0.35 ) ( 0.73 ) ( 0.22 )
AIND 0.147072 -0.00553 -0.02517 5.088743*
( 0.88 ) ( -0.58 ) ( -0.48 ) ( 1.83 )
LEV -1.85835*** -0.1917*** 0.014301 -35.8576***
( -7.76 ) ( -13.97 ) ( 0.19 ) ( -9.01 )
-0.03251**
-0.05158 -0.08047*** 0.511389
LOG_TA *
( -0.55 ) ( -6.07 ) ( -2.71 ) ( 0.33 )
-0.87615**
0.115476*** -0.00055 -0.00194
FAGE *
( 8.8 ) ( -0.74 ) ( -0.47 ) ( -4.02 )
INT_INV 0.018833*** 0.000324 -0.00202* 0.257011***
( 4.92 ) ( 1.48 ) ( -1.67 ) ( 4.04 )

Observations 2688 2688 2688 2688

R-squared 0.7344 0.6113 0.2538 0.4324
Prob (F-
<0.0001 <0.0001 <0.0001 <0.0001
statistic)
Regression results for the fixed effects models are reported above. The table also shows the number of
observations used for the analysis, R –squared statistic. In addition, Prob (F-statisitc) reports the p-
value to test the null hypothesis for no fixed effects. t-statistic is reported in parenthesis.
***, ** and * denotes significance at 1%, 5% and 10% respectively

Table V. Regression analysis using fixed effects model (with pressure-insensitive

institutional ownership and pressure-sensitive institutional ownership variables)

Independent Model 1 Model 2 Model 3 Model 4

Variables (TQ) (ROA) (ROE) (NPM)
Intercept -3.24146*** 0.627765*** 1.302233*** 54.91738***
( -5.37 ) ( 17.98 ) ( 6.74 ) ( 5.42 )
BSIZE 0.038471*** -0.00058 0.003502 -0.19387
( 3.01 ) ( -0.79 ) ( 0.86 ) ( -0.90 )
BIND -0.46936* -0.02743* -0.09724 -5.20698
( -1.73 ) ( -1.75 ) ( -1.12 ) ( -1.14 )
CEO_DUAL 0.2472 0.029638* 0.165826* 0.390843
( 0.85 ) ( 1.75 ) ( 1.77 ) ( 0.08 )
AMEET 0.007363* -0.00011 0.000778 0.009472
( 1.82 ) ( -0.47 ) ( 0.60 ) ( 0.14 )
AIND 0.1392 -0.00568 -0.02653 5.039209*
( 0.84 ) ( -0.60 ) ( -0.50 ) ( 1.82 )
LEV -1.62531*** -0.1867*** 0.048382 -34.4916
( -6.81 ) ( -13.52 ) ( 0.63 ) ( -8.61 )
-0.08524**
-0.06665 -0.03272*** 0.425127
LOG_TA *
( -0.72 ) ( -6.12 ) ( -2.88 ) ( 0.27 )
FAGE 0.105873*** -0.00077 -0.00279 -0.9375***
( 8.15 ) ( -1.03 ) ( -0.67 ) ( -4.30 )
P_INSENSITIV
0.031033*** 0.000567** -0.00003 0.33235***
E
( 7.53 ) ( 2.38 ) ( -0.02 ) ( 4.80 )
-0.00704**
-0.02774*** -0.00078** -0.00767
P_SENSITIVE *
( -4.09 ) ( -1.99 ) ( -3.24 ) ( -0.07 )
 Observations 2688 2688 2688 2688
R-squared 0.741 0.6128 0.2563 0.4341
Prob (F-statistic) <0.0001 <0.0001 <0.0001 <0.0001
Regression results for the fixed effects models are reported above. The table also shows the number of
observations used for the analysis, R –squared statistic. In addition, Prob (F-statisitc) reports the p-
value to test the null hypothesis for no fixed effects. t-statistic is reported in parenthesis.
***, ** and * denotes significance at 1%, 5% and 10% respectively