DOI : 10.17577/IJERTCONV14IS110141- Open Access

- Authors : Ananth S Hoskere, Dr. Prathima Bhat
- Paper ID : IJERTCONV14IS110141
- Volume & Issue : Volume 14, Issue 11, CONFLUENCE – 2026
- Published (First Online) : 20-11-2026
- ISSN (Online) : 2278-0181
- Publisher Name : IJERT
- License:
This work is licensed under a Creative Commons Attribution 4.0 International License
Corporate Governance and Financial Performance: Evidence from Listed Indian SMEs
Author: Ananth S Hoskere, Assistant Professor (Adjunct), School of Business and Management, Christ University, Bangalore. PhD Research Scholar, Department of MBA, BMS College of Engineering, Visvesvaraya Technological University.
Email : ananth.hoskere@christuniversity.in
Co-Author : Dr Prathima Bhat, Professor, Dept. of MBA, BMS College of Engineering, Bangalore.
Abstract – This study examines whether promoter ownership and indirect disclosure measures are associated with financial performance among 115 Indian firms listed through the BSE SME platform over FY 202324 to FY 202526. The source panel contains 345 firm-years. The ownership and cash framework adapts Singh and Rastogi (2023), while SPV is measured as the standard deviation of weekly equity returns following Baumann and Nier (2004), replacing the earlier price-level variance measure. After a verified equity correction and exclusion of undefined ratios, incomplete observations and singleton firms, the six main models use 274344 observations from 96115 firms. ROE and operating profit margin are regressed separately on ownership, weekly return volatility and cash-to-assets, with debt, size and retained earnings relative to equity as controls. Firm and year effects and firm-clustered inference are used. None of the three governance measures is significant at 5% in the main models or reported sensitivity checks. Weekly volatility has negative but imprecise coefficients in both outcomes. Cash-to-assets is weakly significant at 10% only for winsorized operating margin. The findings do not establish a stable governanceperformance relationship or causal effects.
Keywords: Corporate Governance; Ownership Concentration; Information Disclosure; Financial Performance; Listed SMEs; Panel Data.
-
Introduction
Extensive literature on corporate governance (CG) emphasises the relevance of CG for any business firm irrespective of their size and nature. This study aims to examine how corporate governance influence the financial performance in case of Small and Medium Enterprises (SMEs). SME specific studies on CG is limited especially in case of emerging economies (Al-Najjar and Clark, 2017; Hakimah et al., 2019). CG and its impact on FP in case of SMEs is different than in case of large corporations (Saxena and Jagota, 2015). This is mainly because of the differences in the ownership structures (Dasilas and Papasyriopoulos, 2015). For instance, Separation of ownership from management is clearly evident in large corporations, but in SMEs, owners are engaged in managing the firms. Consequently, directly applying corporate governance findings from large firms to listed SMEs may be inappropriate, though these insights can still help frame future research questions for smaller enterprises. While the importance of sound governance in listed SMEs is well-established (Saxena and Jagota, 2015), its specific impact on financial performance remains an empirical research gap. Publicly listing small and medium enterprises on stock exchanges is a relatively recent development in many emerging economies. In India, for example, a dedicated SME listing platforms were launched in 2012 by Bombay Stock Exchange and National Stock Exchange. These provide a sample to test how governance influences financial performance in case of SMEs.
Public listing significantly influences two fundamental dimensions of corporate governance in SMEs: ownership structure and information disclosure. The issuance of common stock inherently alters the ownership concentration of these smaller enterprises. Regulations in capital markets alongside exchange-specific listing and compliance requirements are enabling Information Disclosure practises of SMEs. Ownership concentration and information disclosures are linked to FP in the existing research
literature. Singh and Rastogi, in their paper, have investigated the below research questions : first, is there an effect of ownership concentration on the FP of listed SMEs. Second, how information disclosures of listed SMEs influence their FP. This study is an attempt to extend the study conducted by Singh and Rastogi (2023) by looking at a more recent period. Singh and Rastogi demonstrated that ownership concentration is not significantly related to financial performance and information disclosures are inversely significant for financial performance during the 2018-2020 period. Their findings motivate a later-period assessment, rather than an assumption that the earlier relationship remains unchanged. By testing these hypotheses against a fresh dataset, this study investigates whether those foundational insights remain resilient in post pandemic market landscape.
The present study investigates two questions: whether changes in promoter ownership are positively associated with return on equity (ROE) and operating profit margin (OPM), and whether indirect disclosure proxies display the negative coefficients predicted by a positive disclosure hypothesis. The contribution is an updated three-year assessment using explicit availability rules, year effects and sensitivity tests for difficult accounting denominators and thin trading. The empirical scope is ownership and indirect disclosure measures. It does not estimate effects of board independence or CEO duality.
-
Literature review and hypothesis
Ownership concentration
Concentrated ownership can strengthen monitoring because a large shareholder has both the incentive and the voting power to oversee managerial decisions. In an owner-managed SME, this may reduce conflicts between owners and professional managers. Control can also create conflicts with minority investors when the controlling group obtains private benefits or dominates related-party decisions. Shleifer and Vishny (1997) discuss both the role of large investors in governance and the problems associated with concentrated control. Accordingly, a higher promoter fraction need not have a uniformly beneficial performance association.
Ownership is also potentially endogenous. Profitable firms may attract investors or alter their capital structure, while promoter stakes may respond to financing needs. Demsetz and Villalonga (2001) found no systematic performance relationship when treating ownership as multidimensional and endogenous. In the more directly relevant Indian SME setting, Singh and Rastogi (2023) likewise found insignificant ownership coefficients. The following directional hypothesis tests the monitoring-benefit prediction in the later sample without assuming that the competing explanations are absent.
H1. All else equal, ownership concentration is positively related to financial performance for listed SMEs.
Ownership concentration (OC) is the fraction of equity held by promoters and the promoter group. H1 predicts a positive OC coefficient in each outcome model. An insignificant estimate leaves the directional hypothesis unsupported. It does not establish that ownership has no economic effect.
Information disclosures
Disclosure helps outside investors assess performance and governance when they cannot observe managerial actions directly. Healy and Palepu (2001) describe financial reporting and disclosure as communication mechanisms and examine their capital-market consequences. The relevant empirical challenge is distinguishing disclosure quality from other determinants of observable firm outcomes.
This study retains the inverse-disclosure interpretation of volatility and cash-to-assets (CBTA) used by Singh and Rastogi (2023), but changes the volatility measurement. Baumann and Nier (2004, pp. 33 and 38) define volatility s the standard deviation of weekly equity returns and measure disclosure separately using an index. They provide evidence linking disclosure with lower return volatility; they
do not validate volatility as a direct disclosure score. Here SPV denotes weekly return standard deviation. Business risk, trading conditions and corporate actions can affect volatility, while liquidity needs and investment opportunities affect cash. The inverse-proxy interpretation therefore remains an assumption.
H2. All else equal, information disclosures are positively related to the financial performance of listed SMEs.
H2 predicts negative SPV and CBTA coefficients in both outcome models, conditional on their inverse-proxy interpretation. A negative and significant estimate would be consistent with the hypothesis under that assumption. A positive estimate has the opposite direction, and an insignificant estimate provides insufficient evidence for the predicted association.
-
Research design and methodology
Sample selection and data collection
The supplied BSE listing register records 532 active and 30 suspended firms, giving a sampling frame of 562 active or suspended listings. Of these, 134 were listed by 31 December 2021. Financial observations were assembled for 115 firms across the three financial years ending 31 March 2024, 2025 and 2026. Thus, the financial source panel contains 345 firm-years. This eligibility rule avoids including firms that were first listed during the observation period, but it also limits generalization to more recent listings.
The monetary inputs use standalone financial statements normalized to INR lakh. Assets, cash, borrowings, retained earnings and ownership are documented in a company-year source register using BSE annual reports and year-end shareholding filings (BSE Limited, 2026). The original daily-price file contains positive closes on positive-volume days over 1 April 2023 to 31 March 2026. Its firm-year counts and price variances were reconciled with the supplied workbook before the weekly transformation. Prices are unadjusted for dividends and corporate actions; the resulting returns are closing-price returns rather than total returns.
Nineteen zero-sales firm-years have undefined OPM and are omitted where required. The weekly SPV measure is available for 293 firm-years and missing for 52 with fewer than two valid weekly returns. One GROWTH value remains unavailable because retained-earnings disclosures cannot be reconciled. Missing values are not replaced with zero.
Table 1. Data coverage and estimation samples
Measure or model
Observations
Firms
Financial source panel
345
115
ROE / OPM availability
345 / 326
115 / 113
OC, CBTA, DEBT and SIZE availability
345 each
115
SPV / GROWTH availability
293 / 344
104 / 115
ROE models: OC and CBTA
344 each
115
ROE model: SPV
286
98
OPM models: OC and CBTA
323 each
111
OPM model: SPV
274
96
Common estimation sample for all six models
274
96
Measure or model
Observations
Firms
Balanced common sample
246
82
Source: authors calculations. Model samples exclude missing required variables and firms with only one usable year. The source panel is balanced in its firm-year structure; the full estimation panel is incomplete and unbalanced. The 280 common complete cases reduce to 274 after removing six singleton firms. The balanced sample has 82 × 3 = 246 firm-years.
Description of variables
Table 2 states the measurement conventions. SPV follows Baumann and Niers (2004) weekly-return standard-deviation definition rather than the 252-times-price-variance formula printed by Singh and Rastogi (2023). The source does not specify the Bloomberg return construction or divisor in detail. This implementation uses simple closing-price returns and sample standard deviation, with no annualization or million-unit scaling. Other ratios are fractions. The literal GROWTH formula simplifies to cumulative retained earnings divided by closing equity when PAT is nonzero; it is not a conventional sustainable growth rate.
Table 2. Definitions of dependent variables and regressors
Variable
Measurement
Interpretation
ROE
Profit after tax / closing equity
Financial performance
OPM
EBIT / sales
Financial performance
OC
Promoter and promoter-group shares / total equity shares
Ownership concentration
SPV
Sample standard deviation of weekly simple closing-price returns; at least two returns
Return volatility; conditional inverse disclosure proxy
CBTA
Cash and identified bank deposits / total assets
Inverse disclosure proxy; also liquidity
DEBT
Borrowings / (closing equity + borrowings)
Financing control; leases excluded
SIZE
Natural logarithm of total assets measured in INR
Firm-size control
GROWTH
(Cumulative retained earnings / PAT) × ROE
Retained earnings / equity when defined
Ownership and accounting definitions adapt Singh and Rastogi (2023) to the documented collection conventions. EBIT is earnings before interest and tax; PAT is profit after tax. Undefined denominators and unresolved inputs are omitted.
Weekly observations use the last positive-volume close in each Friday-ending week. A return is P_w / P_(w1) 1 only when both consecutive weeks have observed prices; gaps are not bridged or filled. Returns are assigned to the financial year of the Friday endpoint. A prior-year endpoint is used when available, and partial weeks beyond 31 March 2026 are omitted. SPV is the sample SD of at least two returns, using divisor n1.
Model specifications and econometric tools
Six separate models reproduce the core ownership/disclosure structure of the reference study, with the addition of financial-year effects. Each outcome is regressed on one governance measure at a time,
together with DEBT, SIZE and GROWTH. Firm effects absorb persistent differences such as business model and managerial characteristics. Year effects capture common annual shifts. The estimating equation is:
= + + + 1 + 2 + 3 +
Here Y is ROE or OPM, G is OC, weekly return standard deviation (SPV), or CBTA; i identifies the firm and t the financial year. Models M1M3 use ROE with OC, SPV and CBTA respectively; M4M6 use OPM in the same order. SPV is a fraction: 0.10 represents a weekly return standard deviation of 10%. Fixed-effects slopes are identified by within-firm variation. Firms with only one usable observation are excluded from the corresponding model.
Standard errors are clustered by firm with finite-sample corrections, allowing heteroskedasticity and dependence between observations of the sam firm (Cameron and Miller, 2015). Reported coefficient p- values use a two-sided t reference distribution with the number of contributing firms minus one degrees of freedom. Models use available cases without imputation. Baseline data are not winsorized; the influence of extreme values is examined separately. Panel OLS estimates were independently checked against least-squares regressions with explicit firm and year indicators.
-
Data analysis and econometric results
Descriptive statistics
Table 3. Descriptive statistics before winsorization
Variable
n
Mean
SD
Median
Minimum
Maximum
ROE
345
0.0440
2.0593
0.0249
-29.7177
22.3264
OPM
326
-0.1254
5.7877
0.0776
-75.0118
61.0971
OC
345
0.5189
0.2138
0.5787
0.0000
0.9500
SPV
293
0.0971
0.0371
0.0949
0.0000
0.2434
CBTA
345
0.0881
0.1539
0.0243
0.0000
0.9010
DEBT
345
0.1183
1.4786
0.0956
-25.8534
1.6300
SIZE
345
19.5039
1.0211
19.5031
16.3397
22.0743
GROWTH
344
-0.8157
10.2550
0.1077
-125.4045
43.0224
Source: authors calculations after data corrections. SPV and the accounting ratios are fractions; SIZE is a logarithm. SD is the sample standard deviation. Differences in n reflect variable-specific availability.
Mean promoter ownership is 51.89%, and the median is 57.87%, indicating substantial concentration in the sampled firms. Median ROE is 2.49%, while median OPM is 7.76%. The negative mean OPM of 12.54% contrasts with its positive median because a small number of extreme losses relative to sales affect the mean. ROE ranges from 29.7177 to 22.3264, and OPM ranges from 75.0118 to 61.0971. These are ratios, not percentage-point values. Small denominators and negative equity complicate comparisons, supporting the use of sensitivity analysis rather than treating the mean as representative.
The median source firm-year provides 29 usable weekly returns. Mean SPV is 0.0971 and its median is 0.0949, corresponding to weekly return standard deviations of 9.71% and 9.49%. SPV ranges from 0 to
0.2434. Sparse observations and price jumps associated with corporate actions may affect these estimates. A separate restriction to at least 26 weekly returns assesses sensitivity to limited weekly coverage.
Correlation matrix and VIFs
Table 4. Regressor correlations and variance inflation factors
Variable
OC
SPV
CBTA
DEBT
SIZE
GROWT H
VIF
OC
1.000
1.12
SPV
0.009
1.000
1.06
CBTA
0.263
-0.088
1.000
1.24
DEBT
-0.020
0.082
-0.008
1.000
1.03
SIZE
0.088
-0.149
-0.296
0.105
1.000
1.21
GROWTH
-0.042
0.032
0.010
-0.071
0.061
1.000
1.02
Source: authors calculations using 280 common complete cases before singleton removal. Pearson correlations are reported. SPV is weekly return standard deviation. VIFs use the six regressors jointly with an intercept, rather than the separate model designs.
Pooled VIFs range from 1.02 to 1.24; after removing firm and year indicators, they range from 1.02 to
1.07. These checks show little linear redundancy among observed regressors, but do not resolve omitted-variable bias or validate disclosure proxies. Promoter ownership changes within 78 of the 115 firms, so ownership estimates draw on less variation than the pooled ownership distribution suggests.
Regression analysis
Tables 5 and 6 report the re-estimated models. No governance coefficient is significant at 5%. Ownership is negative for ROE and positive for OPM (p = 0.530 and 0.794). Weekly SPV is negative for both ROE (coefficient 3.2212, p = 0.231) and OPM (12.5818, p = 0.265), matching the direction predicted by H2 but with substantial uncertainty. A 0.01 increase in weekly return standard deviation corresponds to estimated changes of 0.0322 in ROE and 0.1258 in OPM, conditional on the model; these are imprecise associations rather than causal effects. CBTA is also negative and insignificant. Baseline control estimates are imprecise.
Table 5. Firm and year fixed effects with ROE as the outcome
Regressor or statistic
M1 (OC)
M2 (SPV)
M3 (CBTA)
Governance proxy
-0.5365 (0.8523)
p = 0.530
-3.2212 (2.6732)
p = 0.231
-0.0349 (1.6072)
p = 0.983
DEBT
-0.0065 (0.0295)
p = 0.826
-0.7751 (1.4134)
p = 0.585
-0.0065 (0.0296)
p = 0.826
SIZE
0.1832 (0.5491)
p = 0.739
0.1284 (0.2722)
p = 0.638
0.2084 (0.5252)
p = 0.692
GROWTH
0.0349 (0.1582)
p = 0.826
0.0372 (0.1725)
p = 0.830
0.0350 (0.1585)
p = 0.825
Observations
344
286
344
Firms / clusters
115
98
115
Within R²
0.0301
0.0329
0.0299
Source: authors calculations. Cells give coefficient, firm-clustered standard error in parentheses, and coefficient p-value. Each model includes the governance proxy identified in its column heading. SPV is weekly return SD as a fraction, without annualization. Within R² is the fit after demeaning by firm and is distinct from fit after removing both firm and year effects.
Table 6. Firm and year fixed effects with OPM as the outcome
Regressor or statistic
M4 (OC)
M5 (SPV)
M6 (CBTA)
Governance proxy
0.7049 (2.6962)
p = 0.794
-12.5818 (11.2259)
p = 0.265
-7.5849 (8.4680)
p = 0.372
DEBT
-0.0690 (0.0609)
p = 0.260
-0.6686 (1.6100)
p = 0.679
-0.0704 (0.0624)
p = 0.261
SIZE
2.2105 (2.3017)
p = 0.339
2.1547 (2.3740)
p = 0.366
2.2932 (2.3397)
p = 0.329
GROWTH
0.0009 (0.0065)
p = 0.891
0.0048 (0.0089)
p = 0.591
0.0014 (0.0066)
p = 0.838
Observations
323
274
323
Firms / clusters
111
96
111
Within R²
0.0056
0.0060
0.0181
Source: authors calculations. Cells give coefficient, firm-clustered standard error in parentheses, and coefficient p-value. Each model includes the governance proxy identified in its column heading. SPV is weekly return SD as a fraction, without annualization. Within R² is the fit after demeaning by firm and is distinct from fit after removing both firm and year effects.
Diagnostics and endogeneity tests
Conventional joint tests of firm and year effects give p = 0.045 and 0.046 for the ROE ownership and cash models, p = 0.095 for the ROE weekly-SPV model, and p < 0.001 for all three OPM models. These are descriptive tests with restrictive error assumptions. Firm and year effects are retained for all models on design grounds to control persistent firm differences and common annual changes.
Table 7. Cluster-robust tests of firm-mean regressors
Model
F statistic
Degrees of freedom
p-value
ROE_OC
0.840
4, 114
0.502
ROE_SPV
0.822
4, 97
0.514
ROE_CBTA
0.312
4, 114
0.870
OPM_OC
0.707
4, 110
0.589
OPM_SPV
0.328
4, 95
0.858
OPM_CBTA
0.940
4, 110
0.444
Source: authors calculations. Mundlak-type pooled regressions include the model regressors, their firm-specific sample means and year indicators; covariance is clustered by firm. The null is that all four firm-mean coefficients equal zero (Mundlak, 1978). These tests concern correlation with persistent heterogeneity, not all forms of endogeneity.
The firm-mean tests in Table 7 do not reject the null in any model. They therefore do not independently establish a statistical preference for fixed effects, and nonrejection is not proof that firm heterogeneity is unrelated to the regressors. Clustered covariance is used regardless of preliminary heteroskedasticity or serial-correlation tests. With only three years, unit-root procedures and richer dynamic specifications would add little dependable evidence. Common residual dependence across firms may remain despite year effects; three time clusters are inadequate for dependable two-way cluster inference.
No valid external instrument or credible exclusion restriction is available for ownership or disclosure in this design. Accordingly, no instrumental-variable endogeneity test is claimed, and a conventional DurbinWuHausman test cannot establish exogeneity here. Reverse causation and time-varying omitted factors remain possible. In particular, ROE and GROWTH share closing equity in their denominators. Fixed effects and clustered standard errors do not remove this mechanical relationship or establish causality.
Robustness analysis
Sensitivity checks use the 274-observation common estimation sample, the balanced subsample of 82 firms observed in all three years, 1st/99th percentile caps, positive-equity observations, models excluding GROWTH, and restrictions to at least 60 traded days or 26 valid weekly returns per firm- year. Cutoffs are sensitivity choices, not exchange requirements. Percentile limits are calculated over each variables available corrected-panel values before model-specific omission. Singleton firms are removed after each restriction.
Table 8. Sensitivity of governance coefficients
Check
Outcome
OC
SPV
CBTA
n by model
Common sample
ROE
-0.5558
(p = 0.468)
-3.4612
(p = 0.235)
-0.1170
(p = 0.923)
274 / 274 / 274
Common sample
OPM
1.0230
(p = 0.756)
-12.5818
(p = 0.265)
-9.1092
(p = 0.360)
274 / 274 / 274
Balanced common sample
ROE
-0.4823
(p = 0.592)
-3.9842
(p = 0.248)
-0.1373
(p = 0.910)
246 / 246 / 246
Balanced common sample
OPM
0.2098
(p = 0.930)
-4.4754
(p = 0.328)
-1.3990
(p = 0.586)
246 / 246 / 246
1%/99%
winsorized
ROE
-0.4153
(p = 0.283)
-0.1595
(p = 0.839)
-0.1096
(p = 0.802)
344 / 286 / 344
1%/99%
winsorized
OPM
-0.3887
(p = 0.802)
-3.3124
(p = 0.274)
-2.6457
(p = 0.092)
323 / 274 / 323
Positive equity
ROE
-0.1359
(p = 0.690)
-2.2066
(p = 0.328)
1.2362
(p = 0.302)
332 / 278 / 332
Positive equity
OPM
0.2835
(p = 0.923)
-12.6168
(p = 0.259)
-7.9396
(p = 0.374)
312 / 266 / 312
Without GROWTH
ROE
-0.7163
(p = 0.144)
-1.8295
(p = 0.780)
0.3425
(p = 0.724)
345 / 287 / 345
Without GROWTH
OPM
0.7183
(p = 0.790)
-12.2387
(p = 0.277)
-7.5860
(p = 0.370)
324 / 275 / 324
60 traded days
ROE
-0.4912
(p = 0.462)
-0.0950
(p = 0.929)
-0.9529
(p = 0.193)
194 / 194 / 194
60 traded days
OPM
1.6661
(p = 0.436)
-5.5181
(p = 0.331)
-4.4859
(p = 0.203)
185 / 185 / 185
26 weekly returns
ROE
0.1237
(p = 0.590)
0.0319
(p = 0.937)
-0.0523
(p = 0.874)
175 / 175 / 175
26 weekly returns
OPM
1.7148
(p = 0.517)
-8.3844
(p = 0.236)
-5.5386
(p = 0.138)
166 / 166 / 166
Source: authors calclations. Cells show coefficients and two-sided firm-clustered p-values. Observation counts follow the OC / SPV / CBTA model order. Every model includes firm and year effects; controls match the baseline except where GROWTH is explicitly omitted.
No governance coefficient reaches 5% significance in Table 8. Winsorized CBTAOPM is 2.6457 (p
= 0.092), providing only weak 10% evidence; its baseline is 7.5849 (p = 0.372). As a return- construction check, replacing simple-return SD with the sample SD of weekly log returns yields coefficients of 1.0951 for ROE (p = 0.332; n = 286) and 4.3376 for OPM (p = 0.376; n = 274). The 26-week restriction and log-return checks do not establish robustness to dividends or corporate actions, which require adjusted prices.
Econometric results
H1 remains unsupported in both outcomes: the ownership estimates do not establish a positive association. H2 is unsupported at 5% because both weekly-SPV and CBTA coefficients, although negative as predicted, are imprecise. For a 0.01 increase in SPV, the 95% interval for the fitted ROE change is -0.0853 to 0.0208 and for OPM it is -0.3487 to 0.0970. These intervals encompass zero and substantively different effects. Non-significance does not demonstrate that governance or disclosure lacks value.
-
Discussion on findings
The findings do not establish a statistically significant relationship between the selected governance measures and financial performance among Indian listed SMEs during FY 202324 to FY 202526. None of the ownership concentration (OC), weekly return volatility (SPV) or cash-to-assets (CBTA) coefficients is significant at the 5% level for either return on equity (ROE) or operating profit margin (OPM). Accordingly, H1 and H2 remain unsupported within the studys empirical framework.
Despite substantial promoter controlmean ownership of 51.89% and median ownership of 57.87% the ownership estimates provide no consistent evidence of improved profitability. OC is negatively associated with ROE (0.5365; p = 0.530) and positively associated with OPM (0.7049; p = 0.794), but neither estimate is precise enough to support a directional conclusion. These findings resemble the insignificant ownership relationship reported by Singh and Rastogi (2023). Concentrated ownership may strengthen monitoring while also creating conflicts between controlling and minority shareholders, as discussed by Shleifer and Vishny (1997); however, the present analysis cannot distinguish these mechanisms. Moreover, fixed-effects estimates capture changes within firms rather than differences between firms with persistently high and low promoter ownership. Ownership changes occur in only 78 of the 115 firms, limiting the variation available over three years.
The baseline SPV coefficients are negative for both ROE (3.2212; p = 0.231) and OPM (12.5818; p
= 0.265). Their direction is consistent with H2 only if lower volatility reflects better disclosure. A one- percentage-point increase in weekly return standard deviation corresponds to estimated declines of 3.22 percentage points in ROE and 12.58 percentage points in OPM. However, the respective 95% confidence intervals range from 8.53 to 2.08 and from 34.87 to 9.70 percentage points. These wide intervals encompass zero and substantially different associations, preventing a firm conclusion about economic magnitude. Volatility also reflects business risk, trading conditions and corporate actions, so its negative coefficient cannot independently establish a disclosure effect.
CBTA likewise has negative, insignificant baseline coefficients for ROE (0.0349; p = 0.983) and OPM (7.5849; p = 0.372). Cash balances reflect liquidity requirements and investment opportunities as well as possible governance influences, making their interpretation as an inverse disclosure measure conditional. Consequently, the results concern indirect proxies rather than disclosure quality itself.
The sensitivity checks reinforce the absence of strong statistical support: no governance coefficient reaches 5% significance across the common and balanced samples, winsorization, positive-equity restrictions, exclusion of GROWTH or trading-coverage restrictions. Nevertheless, some magnitudes and signs change, indicating that the baseline estimates should not be regarded as stable economic effects. Winsorized CBTAOPM provides weak evidence at 10% (2.6457; p = 0.092), but this isolated result across multiple related specifications does not establish a dependable disclosureperformance relationship. Extreme accounting denominators, incomplete price coverage and sample selection remain relevant sources of uncertainty.
For managers and investors, the evidence does not justify increasing promoter concentration or altering cash balances as proven routes to higher profitability. Assessment should instead consider the substance of monitoring, minority-shareholder protection and reporting practices. The study contributes a later-
period assessment of listed SMEs, while its three-year observational design, indirect disclosure measures and unadjusted returns constrain interpretation. Longer panels, direct disclosure scoring and adjusted price data would improve measurement and precision; stronger identification would be required to establish causality.
-
Conclusion
This study examined the association between promoter ownership, indirect disclosure measures and financial performance among 115 Indian listed SMEs during FY 202324 to FY 202526. Firm and year fixed-effects estimates provide no statistically significant evidence at the 5% level that ownership concentration, weekly return volatility or cash-to-assets is associated with either return on equity or operating profit margin. Neither hypothesis is therefore supported within the studys empirical framework. Although the baseline disclosure- proxy coefficients are negative, their imprecision and sensitivity to alternative specifications prevent a firm conclusion about the magnitude or stability of the relationship. The weak cash- to-assets association with operating margin after winsorization is insufficient to alter this conclusion.
The study contributes a later-period assessment of governance and performance in listed Indian SMEs, while demonstrating the interpretive limitations of ownership ratios and indirect disclosure measures. The findings should not be understood as evidence that governance is economically unimportant. Rather, the three-year observation period, incomplete trading coverage and extreme accounting denominators constrain what the available data can establish. The observational design also leaves reverse causation and time-varying omitted factors unresolved.
These limitations identify several priorities for future research. First, a longer panel would provide more within-firm variation and permit investigation of whether governance changes precede subsequent performance changes. Extending coverage to newer listings and other SME platforms would also help assess the generalizability of the findings. Second, future studies should measure disclosure directly through a systematically constructed index of reporting completeness, timeliness and transparency. Examining such an index alongside volatility and cash holdings would help distinguish disclosure quality from business risk, liquidity requirements and trading conditions.
Third, research should incorporate governance characteristics beyond promoter ownership, including board independence, ownershipmanagement overlap and related-party oversight. These measures could help distinguish the monitoring benefits of concentrated control from potential conflicts with minority shareholders. Longer and broader samples would also permit tests of nonlinear ownership relationships and differences across industries, firm maturity and financing constraints.
Finally, future analysis should use dividend- and corporate-acton-adjusted returns, examine the consequences of missing trading observations and assess profitability measures less sensitive to small or negative denominators. Credible policy changes or other defensible identification strategies would be needed to move from conditional associations towards causal conclusions. Together, these extensions would clarify which governance practices influence SME performance, through what mechanisms and under which conditions.
References
Al-Najjar, B. and Clark, E. (2017), Corporate governance and cash holdings in MENA: evidence from internal and external governance practices, Research in International Business and Finance, Vol. 39,
pp. 1-12.
Baumann, U., & Nier, E. (2004). Disclosure, volatility, and transparency: An empirical investigation into the value of bank disclosure. Economic Policy Review, 10(2), 3145.
Cameron, A. C., & Miller, D. L. (2015). A practitioners guide to cluster-robust inference. Journal of Human Resources, 50(2), 317372.
Dasilas, A. and Papasyriopoulos, N. (2015), Corporate governance, credit ratings and the capital structure of Greek SME and large listed firms, Small Business Economics, Vol. 45 No. 1, pp. 215-244.
Demsetz, H., & Villalonga, B. (2001). Ownership structure and corporate performance. Journal of Corporate Finance, 7(3), 209233.
Hakimah, Y., Pratama, I., Fitri, H., Ganatri, M. and Sulbahri, R.A. (2019), Impact of intrinsic corporate governance on financial performance of Indonesian SMEs, International Journal of Innovation, Creativity and Change, Vol. 7 No. 1, pp. 32-51.
Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature. Journal of Accounting and Economics, 31(1 3), 405440.
Mundlak, Y. (1978). On the pooling of time series and cross section data. Econometrica, 46(1), 6985.
Saxena, A. and Jagota, R. (2015), Should the MSMEs be governed the corporate way?, Indian Journal of Corporate Governance, Vol. 8 No. 1, pp. 54-67.
Shleifer, A., & Vishny, R. W. (1997). A survey of corporate governance. Journal of Finance, 52(2), 737 783.
Singh, K., & Rastogi, S. (2023). Corporate governance and financial performance: Evidence from listed SMEs in India. Benchmarking: An International Journal, 30(4), 14001423. First published online 13 June 2022.
