Raju G and Riny Raju, Financial Distress Prediction in Indian Small-Cap Companies: A Longitudinal Analysis Using Altman’s Z-Score Models (2025)

Summary

A 2025 journal article by Raju G of the Xavier Institute of Management and Entrepreneurship and Riny Raju of the University of Kerala, Financial Distress Prediction in Indian Small-Cap Companies: A Longitudinal Analysis Using Altman's Z-Score Models, in International Insurance Law Review, Vol 33 S5 (2025), pages 551-569. The abstract says the study applies Altman's Z-score and Z''-score models to 30 NSE listed companies observed semi-annually from March 2019 to March 2025. It reports Z-score classifications of 60.5 percent Safe, 18.5 percent Grey Zone and 21.0 percent Distressed, and a pandemic peak of 36.7 percent distressed in March 2020 followed by recovery. It also reports differences by sector, with FMCG at 0 percent distressed and Power at 100 percent.

Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used

Cite this document

“Raju G and Riny Raju, Financial Distress Prediction in Indian Small-Cap Companies: A Longitudinal Analysis Using Altman’s Z-Score Models (2025).” Original at lumarpub.com. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/raju-g-and-riny-raju-financial-distress-prediction-in-indian-small-cap-1b0884d020/, SHA-256 fcaa9b30d7bd6f02e96e7ce7e9107217165b0af36b1a21c37ea34a389a5a2838.

Full text

International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models

Raju G1, Riny Raju2
1Professor, Xavier Institute of Management and Entrepreneurship (XIME), Bangalore, India

2Research Scholar, Department of Economics, Government College for Women

(University of Kerala), Thiruvananthapuram, Kerala
Email: 1drrajug1@gmail.com; 2rinyraju1@gmail.com
ORCID: https://orcid.org/0000-0002-3242-3863; 2https://orcid.org/0009-0004-1698-3555
1

Abstract
This study examines financial distress dynamics in Indian small-cap firms using Altman's Z-score
and Z''-score models through longitudinal analysis of 30 NSE listed companies observed semi-
annually from March 2019 to March 2025. The research addresses critical gaps in emerging
market distress prediction literature while capturing the complete COVID-19 pandemic
trajectory. Results reveal significant heterogeneity in small-cap financial health: Z-score
classifications show 60.5 per cent Safe, 18.5 per cent Grey Zone, and 21.0 per cent Distressed
firms, while Z''-score identifies fewer distressed companies (14.9 per cent), yielding moderate
inter-model agreement (Cohen's κ = 0.493). The Z-score exhibits 3.8-fold greater temporal
volatility than Z''-score due to sales-to-assets ratio sensitivity.
Temporal analysis demonstrates a pronounced V-shaped recovery pattern, from pre-pandemic
vulnerability (28.3 per cent distressed) through pandemic crisis peak (36.7 per cent distressed,
March 2020) to robust recovery with 61.7 per cent improvement by March 2025. Sectoral
analysis reveals statistically significant heterogeneity (ANOVA: F = 15.87, p < 0.001), with
defensive sectors (FMCG: 0 per cent; Consumer Durables: 3.8 per cent) contrasting sharply with
infrastructure sectors (Power: 100 per cent; Telecommunications: 76.9 per cent). Sector
classification explains 32 per cent of distress variance. Findings advance bankruptcy prediction
theory through emerging market validation and inform practice via dual-model investment
screening, semi-annual credit monitoring and sector differentiated risk assessment. The study
challenges conventional small firm fragility assumptions, demonstrating substantial adaptive
capacity and resilience among survivors.
Keywords: Financial distress prediction, Altman Z-score, Small-cap companies, Longitudinal
analysis, Bankruptcy prediction
JEL Classification: G32, G33, M41, C38

1. Introduction
Financial distress prediction is crucial in emerging economies like India, where capital market
volatility, institutional uncertainties and structural vulnerabilities threaten firm survival.
Accurately forecasting financial deterioration enables stakeholders—investors, creditors,
regulators and management—to intervene early, preventing wealth destruction and maintaining
systemic stability. This has become particularly important following high profile corporate
failures and India's increasing integration with global financial markets.
Small-cap enterprises comprise approximately 85 per cent of listed entities on Indian stock
exchanges and contribute significantly to employment, innovation and economic growth.
551
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
However, they face unique vulnerabilities: limited capital access, high debt-to-equity ratios, weak
supply chain bargaining power, and heightened sensitivity to macroeconomic shocks. The COVID-
19 pandemic during 2020-2021 exposed these fragilities through widespread liquidity crises and
operational disruptions, highlighting the need for robust distress assessment mechanisms.
Altman's Z-score model has proven remarkably durable over five decades due to its analytical
simplicity, interpretive clarity and empirical validation. Originally developed by Edward Altman
in 1968 using multiple discriminant analysis, the model combines five financial ratios into a
composite score that classifies firms into distinct financial health zones. The Z''-score variant is
particularly relevant for this research, optimized for non-manufacturing and emerging market
contexts by removing the sales-to-assets ratio and recalibrating coefficients. While recognizing
that bankruptcy prediction involves qualitative factors like management quality, governance and
competitive positioning, this study focuses on quantifiable metrics that enable systematic
measurement and temporal comparison. This approach aligns with Altman's foundational
principles while facilitating objective, replicable analysis.
This longitudinal investigation spanning March 2019 through March 2025, covering both normal
economic conditions and pandemic disruption, advances understanding of small-cap financial
distress dynamics in Indian markets while informing enhanced risk assessment frameworks for
emerging economies.
The remainder of this paper is organized as follows. Section 2 presents the theoretical
foundations of financial distress prediction and the evolution of Altman's Z-score models. Section
3 reviews the extensive literature on bankruptcy prediction models, with emphasis on emerging
market applications. Section 4 identifies gaps in existing research that this study addresses.
Section 5 articulates the specific research objectives and questions. Section 6 details the research
methodology, including sample selection, data collection and analytical framework. Section 7
presents the empirical findings across multiple dimensions of analysis. Section 8 concludes with
a synthesis of contributions, practical implications, limitations, and directions for future research.

2. Theoretical Foundations and Evolution of Z-Score Models
Financial distress prediction has emerged as a critical subdomain within corporate finance,
focusing on identifying early warning signals preceding corporate failure. The theoretical
foundations draw from financial ratio analysis, statistical discrimination techniques and market
based valuation theories, operating on the fundamental premise that deteriorating financial
health manifests through observable patterns in financial statements well before actual
bankruptcy events. Financial distress encompasses a spectrum beyond legal bankruptcy,
including persistent losses, loan defaults, debt restructuring, severe liquidity constraints, and
inability to meet obligations. Beaver's pioneering 1966 research established that financial ratios
exhibit predictable deterioration patterns preceding failure, with cash flow ratios demonstrating
particular predictive power, laying groundwork for subsequent multivariate models.
Edward Altman's 1968 Z-score model represents a watershed in bankruptcy prediction research.
Using discriminant analysis on 66 matched firms from 1946-1965, Altman identified five financial
ratios that effectively distinguished between failing and non-failing firms: working capital to total
assets (X₁), retained earnings to total assets (X₂), EBIT to total assets (X₃), market value equity to
book value liabilities (X₄), and sales to total assets (X₅), with weights of 1.2, 1.4, 3.3, 0.6, and 1.0

552
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
respectively. The model achieved 95 per cent accuracy one year before bankruptcy and 72 per
cent two years prior.
Recognizing the original model's limitations to publicly traded manufacturing firms, Altman
developed two variants. The Z'-score (1983) adapted for private manufacturing firms by
replacing market with book value of equity. The Z''-score eliminated the potentially problematic
sales-to-assets ratio and modified coefficients (6.56X₁ + 3.26X₂ + 6.72X₃ + 1.05X₄), making it
particularly suitable for non-manufacturing sectors and emerging markets where sales ratios can
be misleading or volatile.

3. Literature Review
This literature review traces the evolution and application of Altman's Z-score model from its
original formulation through contemporary adaptations, examining its predictive accuracy across
diverse geographical contexts, time periods, and economic conditions with particular focus on
emerging market applications.
Altman (1968) introduced the Z-score model using multiple discriminant analysis to predict
corporate bankruptcy, combining five financial ratios into a single score that achieved 95 per cent
accuracy one year before failure and 72 per cent two years prior. The research established that
multivariate analysis significantly outperforms univariate ratio analysis for bankruptcy
prediction.
Altman (2000) examined the Z-score model's performance over three decades across different
time periods, countries, and economic conditions. While accuracy declined somewhat over time,
the model maintained predictive validity with appropriate threshold adjustments. The study
introduced sector-specific modifications and the Z''-score for emerging markets.
Eljelly and Mansour (2001) applied the Z-score to Sudan's developing economy, finding that while
the model maintained directional validity, optimal cut-off thresholds required local adjustments.
The study documented higher baseline distress rates and greater volatility in emerging markets
compared to developed markets.
Grice and Ingram (2001) tested whether Z-score coefficients from 1960s data remained valid for
1990s bankruptcies, finding significant coefficient drift over time. Re-estimated models improved
accuracy, and optimal cut-offs shifted lower in later periods, possibly reflecting changing
accounting practices or bankruptcy law reforms.
Charitou, Neophytou, and Charalambous (2004) applied multiple bankruptcy prediction models
to UK companies, conducting comprehensive comparative analysis. They found that models
perform differently across industries and time periods, with no single model uniformly superior,
recommending multiple models for robust assessment and emphasizing that model choice
significantly affects classification.
Agarwal and Taffler (2008) compared accounting-based models (Altman and Taffler Z-scores)
with market based models using UK data. Accounting models provided complementary
information, better capturing fundamental financial deterioration while market measures
incorporated forward-looking information and investor sentiment. This suggests the Z-score's
historical basis provides fundamental health assessment distinct from market sentiment.
Lin (2009) applied multiple methodologies to Taiwanese firms, comparing traditional statistical
models with neural networks. Discriminant analysis achieved 80-85 per cent accuracy,

553
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
comparable to sophisticated techniques. Model performance varied significantly across
industries, with service sectors showing lower prediction accuracy than manufacturing.
Rashid and Abbas (2011) examined model validity in Pakistan's South Asian context, finding that
Z''-score outperformed the original Z-score for mixed industry samples, with accuracy improving
through sector-specific threshold adjustments. Family ownership and concentrated shareholding
patterns characteristic of South Asian firms affected distress dynamics.
Bhunia, Mukhuti, and Roy (2011) applied the Z-score to Indian steel industry companies during
2006-2010, including the global financial crisis. They found high inter-firm heterogeneity and
crisis-induced deterioration followed by recovery, recommending the Z-score for Indian
assessment but noting the need for sector-specific benchmarks.
Anjum (2012) conducted meta-analysis of Z-score applications across 45 studies, documenting
that the model maintains moderate to high accuracy (65-85 per cent) across diverse contexts but
requires context specific calibration. Performance deteriorates during crisis periods but recovers
afterward.
Altman et al. (2017) tested the Z-score across 31 European countries, finding that creditor rights
strength, bankruptcy code efficiency and accounting standard quality influence prediction
accuracy, highlighting the need for country specific threshold adjustments.
Korath and Nayak (2022) tested the model on Indian manufacturing companies under the
Insolvency and Bankruptcy Board of India (2017-2021), finding all five unhealthy firms showed
low Z-scores while four of five healthy companies showed higher scores, validating continued
relevance.
Thacker and Saha (2024) applied XGBoost algorithm to Indian manufacturing firms, identifying
common predictive factors including lagged net profit margin, growth of profit after tax, lagged
assets turnover ratio, sales growth, and log of total assets, demonstrating that past performance
significantly influences future distress probability.
Apoorva, Kumar, and Roy (2025) evaluated predictive accuracy for NSE listed companies, finding
the Traditional Altman Z-Score achieved 100 per cent accuracy predicting distress five years prior
for all eight analyzed companies. The Modern and Emerging Market Z-Score models showed only
62.5 per cent accuracy, while the Zmijewski model proved most reliable among alternatives.
While the literature demonstrates extensive validation of bankruptcy prediction models across
diverse contexts, several critical gaps persist in financial distress research, particularly
concerning emerging markets and small-cap enterprises. The following section identifies these
gaps and positions the current study's contribution.

4. Research Gap Analysis
Despite extensive validation of Altman's Z-score model since 1968, critical gaps persist in
financial distress research, particularly concerning emerging markets and small-cap enterprises.
While the Z''-score variant theoretically offers superior performance in non-manufacturing
sectors and developing economies by eliminating the sales-to-assets ratio, comparative empirical
evidence between Z-score and Z''-score remains limited in Indian small-cap contexts, with most
studies applying single model variants without systematic performance comparison.
Existing research predominantly employs cross-sectional designs capturing firm classifications
at specific points, overlooking financial distress's dynamic nature. Limited longitudinal studies
reveal distress involves complex trajectories including deterioration, recovery, and transitions
554
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
between health states. Understanding transition dynamics, persistence versus mobility in
classifications, recovery patterns, grey zone duration and progression patterns, remains
underexplored, especially in emerging markets.
Small-cap firms exhibit unique distress characteristics driven by heightened information
asymmetry, limited capital access, constrained management resources, greater vulnerability to
industry shocks, and distinct governance dynamics through family ownership and concentrated
shareholding. Despite higher baseline distress rates and volatility, research specifically focused
on small-cap segments remains limited.
The COVID-19 pandemic created an unprecedented natural experiment for examining financial
distress through sudden disruptions and extraordinary policy interventions. Studies specifically
tracking small-cap samples through pre-pandemic, crisis and recovery periods remain scarce,
particularly regarding traditional prediction model performance during this shock. Research
employing semi-annual measurement frequency, capturing seasonal patterns and intra-year
dynamics missed by annual assessments, is uncommon. Additionally, sector specific distress
patterns within India's small-cap universe lack comprehensive cross-sectoral comparative
analysis.
This study addresses these gaps through longitudinal analysis of 30 NSE listed small-cap
companies over 13 half-yearly periods from March 2019 to March 2025, employing both Z-score
and Z''-score models with formal concordance analysis, examining cross-sectoral patterns,
capturing the complete COVID-19 cycle, and analyzing transition dynamics between distress
zones. To address these identified gaps comprehensively, this study establishes specific research
objectives and formulates targeted research questions as articulated below.

5. Research Objectives and Questions
This study aims to conduct a comprehensive longitudinal analysis of financial distress patterns in
30 small-cap companies listed on the National Stock Exchange (NSE) of India using both Altman's
Z-score and Z''-score models over a six year period from 31st March 2019 to 31st March 2025.
This timeframe is particularly significant as it encompasses the pre-pandemic period, the COVID-
19 crisis years and the subsequent recovery phase, providing unique insights into how financial
health indicators respond to extraordinary economic shocks.
The primary objective is to examine the temporal evolution of financial health across this
representative sample and identify patterns, trends and determinants of financial distress.
Specifically, this research addresses five interrelated questions:
RQ1: Temporal Evolution of Aggregate Financial Health
How has aggregate financial health evolved over the study period? What proportion of companies
fall within distress, grey and safe zones at different time points?
RQ2: Model Concordance and Divergence
What is the concordance between Z-score and Z''-score classifications? When and why do these
models yield divergent predictions?
RQ3: Seasonal Patterns in Financial Health
Do financial health patterns exhibit seasonal variations between fiscal year-end (March) and mid-
year (September) measurements? What factors explain such variations—fiscal reporting effects,
working capital cycles or business seasonality?

555
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
RQ4: Persistence and Transition Dynamics
How persistent are distress classifications? Do distressed companies show deterioration,
stabilization or recovery patterns across the thirteen observation periods?
RQ5: COVID-19 Impact and Resilience
How did COVID-19 impact financial trajectories? Were companies with stronger pre-pandemic
positions more resilient during the crisis and recovery phases?
These research questions are designed to generate insights with both theoretical significance for
bankruptcy prediction literature and practical value for investment decision making, credit risk
management and policy formulation.
Achievement of these research objectives requires a rigorous methodological approach that
ensures data quality, analytical validity and interpretive reliability. The following sections detail
the research design, sample selection procedures, variable operationalization, and analytical
techniques employed in this investigation.

6. Research Methodology
6.1 Research Design
This study employs a quantitative, longitudinal research design examining financial distress
patterns in Indian small-cap companies over six years. Adopting a positivist epistemological
stance, the research assumes financial distress can be objectively measured through standardized
financial ratios with systematic patterns existing between financial metrics and firm health. The
longitudinal panel data approach enables examination of both cross-sectional variations and
temporal dynamics.
The design incorporates several methodological strengths. First, the extended observation
window from March 2019 to March 2025 captures multiple economic conditions including
normal periods, COVID-19 crisis, and recovery phases. Second, thirteen half-yearly measurement
points balance frequent observation benefits, capturing intra-year dynamics and seasonal
patterns, against stability requirements for meaningful analysis. Third, comparative application
of both Z-score and Z''-score models enables assessment of model specific effects and provides
robustness checks. Fourth, multi-sector sample composition allows examination of industry
specific patterns while maintaining sufficient sample size for statistical validity.

6.2 Population and Sample Selection
This study examines small-cap companies listed on India's National Stock Exchange using
purposive sampling to identify 30 companies satisfying rigorous data requirements. The strategy
emphasizes data integrity and temporal consistency over comprehensive market coverage.
Inclusion criteria include: complete financial data availability across all 13 half-yearly periods
from March 2019 through March 2025; NSE small-cap classification; sectoral diversity ensuring
adequate representation of India's heterogeneous small-cap industrial composition and
Bloomberg Terminal sourced data providing standardized accounting treatment and verified
reliability.
The sample generates 390 firm-period observations (30 companies × 13 periods), providing
substantial data for trend analysis and statistical examination, aligning with precedent
bankruptcy prediction studies. Multi-sector composition enables meaningful cross-sectoral
comparisons while maintaining sufficient observations within major sector categories.
556
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
6.3 Final Sample Composition
The final sample comprises 30 small-cap companies distributed across 11 distinct sectors. This
sector distribution reflects the industrial and manufacturing orientation characteristic of India's
small-cap segment, with substantial representation in capital goods and chemicals sectors that
are central to India's industrialization trajectory. The presence of consumer oriented sectors
(Consumer Durables, FMCG) and defensive sectors alongside cyclical industries enables
examination of how distress patterns vary with business cycle sensitivity.

Table 1: Sample Composition by Sector
Sl. No. Number of
Sector Percentage
Companies
1 Capital Goods 9 30.0
2 Chemicals 6 20.0
3 Consumer Durables 4 13.3
4 Automobile and Auto Components 2 6.7
5 Fast Moving Consumer Goods 2 6.7
6 Power 2 6.7
7 Oil, Gas and Consumable Fuels 1 3.3
8 Metals and Mining 1 3.3
9 Telecommunication 1 3.3
10 Construction Materials 1 3.3
11 Textiles 1 3.3
Total 30 100.0

6.4 Variable Definitions and Calculations
Altman's Original Z-Score Model
The original Z-score model developed by Altman (1968) for publicly traded manufacturing firms
employs the formula:
Z = 1.2X₁ + 1.4X₂ + 3.3X₃ + 0.6X₄ + 1.0X₅
Where X₁ measures short term liquidity (working capital/total assets), X₂ captures cumulative
profitability (retained earnings/total assets), X₃ reflects operating efficiency (EBIT/total assets),
X₄ represents market assessed solvency (market value of equity/book value of total liabilities),
and X₅ indicates asset utilization (sales/total assets). The coefficients reflect each ratio's relative
importance in discriminating between bankrupt and non-bankrupt firms. Threshold values are:
Z > 2.99 suggests financial stability (Safe Zone), Z < 1.81 indicates high distress probability
(Distress Zone), and 1.81 ≤ Z ≤ 2.99 represents a grey zone requiring careful monitoring. The
model balances profitability, liquidity, leverage, and efficiency metrics for comprehensive
financial health assessment.
Altman's Z''-Score Model for Emerging Markets
The Z''-Score (1995) adaptation designed for non-manufacturers and emerging market firms
eliminates the sales-to-assets ratio to address industry specific asset turnover variations and
sales volatility in developing economies:
Z'' = 6.56X₁ + 3.26X₂ + 6.72X₃ + 1.05X₄

557
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
Component variables X₁ through X₄ are calculated identically to the original model. Recalibrated
coefficients reflect increased discriminatory weight assigned to each remaining variable after
excluding X₅. This eliminates concerns about varying asset turnover standards across industries
and potential distortions from sales volatility in emerging markets. Revised classification
thresholds are: Z'' > 2.6 indicates financial stability (Safe Zone), Z'' < 1.1 signals high distress
probability (Distress Zone), and 1.1 ≤ Z'' ≤ 2.6 represents uncertain financial health requiring
careful evaluation (Grey Zone). This specification provides more robust bankruptcy prediction
where asset turnover metrics may be unreliable or incomparable across sectors.

6.5 Analytical Framework
The analytical framework employs multiple complementary techniques. Descriptive statistics
compute central tendency measures (mean, median), dispersion (standard deviation,
interquartile range), and distribution characteristics (skewness, kurtosis) for both models across
all periods. Companies are classified into Safe (Z > 2.99 or Z'' > 2.6), Grey (1.81 ≤ Z ≤ 2.99 or 1.1 ≤
Z'' ≤ 2.6), and Distress (Z < 1.81 or Z'' < 1.1) zones, with distributions tabulated by sector and
period.
Temporal trend analysis tracks mean and median scores across 13 semi-annual periods,
emphasizing changes surrounding COVID-19 onset and recovery phases, while capturing
individual firm trajectory heterogeneity. Paired t-tests compare March versus September
observations to detect systematic seasonal variations attributable to fiscal year-end effects or
working capital cycles.
ANOVA with robust variance estimation tests significant mean score differences across sectors,
identifying specific divergences and industry specific distress propensities. Model concordance
analysis employs concordance matrices and Cohen's kappa statistic to assess Z-score and Z''-
score agreement, examining discordant cases to understand divergence drivers.
Transition matrices tabulate movement frequencies between distress zones across consecutive
periods, revealing persistence rates and trajectories. The sample is segmented into pre-pandemic,
pandemic, and recovery phases, with statistical tests comparing mean scores and zone
distributions. Correlation and regression analyses assess whether pre-crisis strength predicted
pandemic resilience.
Component ratio analysis examines individual ratio components for transitioning firms to
identify primary drivers of classification changes, with sector specific correlations revealing
industry specific financial vulnerabilities.

6.6 Methodological Summary
This research maintains ethical standards by utilizing only publicly available financial data from
listed companies without accessing confidential information or making specific investment
recommendations. Analysis employs standardized, objective methodologies and presents
findings in aggregate form to emphasize patterns rather than singling out individual companies
that could influence market perceptions. The study acknowledges limitations and avoids
overstating model predictive power, ensuring responsible, transparent reporting.
This study employs a longitudinal panel data methodology to examine financial distress dynamics
among Indian small-cap companies over the period 2019-2025. The analysis uses Bloomberg
standardized financial data with parallel application of Altman Z-score and Z''-score models. The
558
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
methodology exhibits four key strengths: extended six-year observation window capturing
complete COVID-19 crisis and recovery phases; semi-annual measurement frequency enabling
detection of intra-year trends and volatility patterns; multi-sector sample composition allowing
identification of industry specific distress patterns; and comprehensive analytical framework
integrating descriptive statistics, temporal analysis, sectoral comparisons, transition matrices,
and pandemic impact assessment. This dual model approach provides robustness checks and
generates insights into optimal distress assessment tools for emerging markets, positioning the
study to contribute meaningfully to understanding financial health evolution in an understudied
segment of India's equity markets with implications for investors, regulators, and corporate
management.
Having established the methodological framework, we now present the empirical findings from
our analysis of 30 small-cap companies across 13 semi-annual observation periods. The results
are organized thematically to address each research question systematically while revealing
interconnected patterns in financial distress dynamics.

7. Empirical Findings
7.1 Descriptive Statistics and Financial Distress Distribution
7.1.1 Comparative Model Distribution Analysis
This analysis examines the distributional patterns and classification concordance between the
original Z-score and Z''-score models to assess their relative performance and identify systematic
differences in distress zone assignments across the sample period.

Table 2: Comparative Descriptive Statistics for Z-Score and Z''-Score (N = 390)
Percentage
Statistic Z-Score Z''-Score Difference
Change
Mean 5.93 4.35 -1.58 -26.6
Median 3.88 4.21 +0.33 +8.5
Standard Deviation 6.05 3.44 -2.61 -43.1
Coefficient of Variation (%) 101.95 79.14 -22.81 -22.4
Minimum -0.34 -4.65 -4.31
Maximum 35.94 26.36 -9.58 -26.7
Range 36.28 31.01 -5.27 -14.5
Q1 (25 percentile)
th 2.15 2.00 -0.15 -7.0
Q3 (75 percentile)
th 8.14 6.38 -1.76 -21.6
Interquartile Range 5.98 4.38 -1.60 -26.8
Skewness 2.23 1.50 -0.73 -32.7
Kurtosis (excess) 6.24 6.68 +0.44 +7.1
Source: Authors' calculations based on data from Bloomberg Terminal (2019-2025)

Zone Classification Distribution Comparison
Z-Score Z''-Score Difference
Classification Zone
Frequency (%) Frequency (%) (%)
Distress 82 58 -6.16

559
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
(21.03%) (14.87%)
72 62
Grey -2.56
(18.46%) (15.90%)
236 270
Safe +8.72
(60.51%) (69.23%)
Source: Authors' calculations based on data from Bloomberg Terminal (2019-2025)

The comparative analysis reveals critical patterns in Indian small-cap financial health
distribution. The finding that 60.5 per cent (Z-score) to 69.2 per cent (Z''-score) of observations
fall in the Safe zone contradicts negative small-cap risk perceptions, indicating most firms
maintain fundamentally sound financial health. This proportion exceeds the 45-50 per cent safe
zone rates in earlier Indian studies, suggesting improved small-cap financial management.
The Z-score distribution's mean (5.93) significantly exceeds its median (3.88), confirmed by
skewness coefficient of 2.23, indicating most companies cluster in moderate ranges while
exceptionally healthy firms elevate the average. The coefficient of variation at 101.95 per cent,
exceeding 100 per cent, signals extreme heterogeneity, suggesting small-cap categorization
masks profound diversity in business models, growth stages, and financial management
sophistication.
The Z''-score demonstrates systematically different distributional properties with 26.6 per cent
lower mean but more stable characteristics. Its coefficient of variation decreases to 79.14 per cent
and skewness reduces to 1.50, suggesting more stable assessments by attenuating outlier
influence. Most strikingly, Z''-score classifies 29.3 per cent fewer observations as distressed
(14.87 per cent vs 21.03 per cent), a substantial difference statistically significant (χ² = 15.7, p <
0.001).
The leptokurtic distributions (excess kurtosis: 6.24 for Z-score, 6.68 for Z''-score) indicate heavy
tails with higher probability of extreme values than normal distributions. Elevated distress rates
(21.0 per cent for Z-score, 14.9 per cent for Z''-score) remain substantially higher than large-cap
samples (typically 10-15 per cent), aligning with theoretical expectations based on small firm
characteristics: limited capital market access during stress, higher operational leverage, and
greater working capital vulnerability. The Grey zone capturing 15-18 per cent represents
companies in genuine uncertainty requiring supplementary qualitative analysis.
These findings align with Pecking Order Theory's predictions about small firm financing
constraints, Information Asymmetry Theory regarding opacity creating wider financial health
spreads, and Life Cycle Theory explaining observed heterogeneity as small caps span multiple
stages from young growth firms to mature declining businesses.

7.2 Temporal Evolution and COVID-19 Impact Analysis
7.2.1 Comparative Period-by-Period Trends
This analysis examines the temporal evolution of financial health indicators across 13 semi-
annual periods from March 2019 to March 2025, tracking how both Z-score and Z''-score metrics
responded to changing economic conditions including the COVID-19 pandemic and subsequent
recovery phases.

560
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
Table 3: Period-by-Period Comparative Statistics
Z''-
Z-Score Z-Score Z''- Z''-
Z-Score Score
Period Date Distress Safe Score Score
Mean Distress
(%) (%) Mean Safe (%)
(%)
1 Mar-25 7.54 6.7 76.7 4.93 10.0 76.7
2 Sep-24 8.80 6.7 83.3 4.83 6.7 76.7
3 Mar-24 6.97 13.3 76.7 4.14 10.0 73.3
4 Sep-23 6.52 13.3 80.0 4.67 6.7 80.0
5 Mar-23 5.07 23.3 60.0 3.79 20.0 63.3
6 Sep-22 5.36 23.3 63.3 4.31 10.0 66.7
7 Mar-22 5.64 23.3 56.7 3.95 20.0 63.3
8 Sep-21 7.16 20.0 66.7 4.83 16.7 76.7
9 Mar-21 6.05 23.3 53.3 4.46 26.7 60.0
10 Sep-20 5.84 26.7 50.0 4.80 20.0 70.0
11 Mar-20 3.86 36.7 40.0 4.01 20.0 63.3
12 Sep-19 3.88 30.0 33.3 4.18 13.3 66.7
13 Mar-19 4.42 26.7 46.7 3.63 13.3 63.3
Source: Authors’ calculations based on data from Bloomberg Terminal (2019-2025)

7.2.2 Three-Phase COVID-19 Impact Assessment
The temporal analysis reveals distinct phases with important implications for small-cap
resilience.
Table 4: Financial Health by COVID-19 Phase
Z- Z''- Z''-
Z- Z-Score Z''-
Score Score Score
Phase Periods N Score Distress Score
Safe Distress Safe
Mean (%) Mean
(%) (%) (%)
Pre- Mar-19,
60 4.15 28.3 40.0 3.90 13.3 65.0
Pandemic Sep-19
Pandemic
Mar-20 30 3.86 36.7 40.0 4.01 20.0 63.3
Onset
Pandemic Sep-20 to
120 6.17 23.3 56.7 4.51 20.8 67.5
Peak Sep-21
Mar-22 to
Recovery 180 6.71 14.5 72.6 4.43 11.2 72.6
Mar-25
Source: Authors’ calculations based on data from Bloomberg Terminal (2019-2025)

Pre-Pandemic Vulnerability (March 2019 to September 2019): The study begins with
moderate financial health showing deterioration before COVID-19. Pre-pandemic weakness (Z-
score mean: 4.15, distress: 28.3 per cent) likely reflects the 2018-2019 NBFC liquidity crisis,
global trade tensions and domestic consumption slowdown. This baseline vulnerability meant

561
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
companies entered COVID-19 from compromised states, potentially explaining the severe March
2020 distress spike.
COVID-19 Impact and Divergent Model Response: March 2020 marks devastating pandemic
arrival with Z-score plummeting to 3.86 (study's lowest) and distress soaring to 36.7 per cent.
However, Z''-score shows remarkably different patterns, minimal mean deterioration (4.01) and
moderate distress increase (20.0 per cent). This divergence suggests the pandemic primarily
affected operational metrics (sales-to-assets ratio in Z-score) rather than fundamental solvency
captured by Z''-score. The Z''-score's stability has important interpretative implications:
eliminating sales-to-assets ratio (X₅) makes it less sensitive to revenue disruptions particularly
acute during lockdowns, suggesting government interventions effectively prevented balance
sheet destruction even as operations struggled.
V-Shaped Recovery Trajectory: Rapid V-shaped recovery, with Z-score mean improving 59.8
per cent from pandemic onset to peak, contradicted expectations of prolonged small firm distress.
This reflects multiple mechanisms: India's comprehensive policy response including ₹3 lakh
crore guaranteed lending and 93 per cent loan moratorium uptake provided critical liquidity;
forced digital transformation created permanent efficiency gains; market consolidation
introduced survivor bias; 7 per cent CPI inflation in 2022 inflated nominal metrics, particularly
affecting Z-score through sales components; and post-lockdown pent-up demand drove sharp
revenue recoveries. The Z-score's 3.8-fold greater temporal volatility (range: 4.94 points vs 1.30
for Z''-score) indicates sales-to-assets ratio exhibits high variability during crises, viewable as
either strength (higher sensitivity for early warning) or weakness (more noise obscuring
solvency signals) depending on analytical objectives.
Sustained Recovery and New Equilibrium: The recovery phase (March 2022 to March 2025)
demonstrates continued strengthening with Z-score reaching 6.71 and Safe zone expanding to
72.6 per cent, an 82 per cent increase from pre-pandemic. Distress drops to 14.5 per cent, a 60
per cent reduction from pandemic onset. Sustained improvement suggests permanent rather
than temporary gains, potentially reflecting accelerated digital transformation and market
consolidation benefits. March 2025 shows slight moderation but maintains substantially stronger
health than pre-pandemic levels, suggesting a new, higher equilibrium for small-cap financial
health post-crisis.

7.3 Model Concordance and Divergence Analysis
This section examines the concordance between Z-score and Z''-score classifications to assess the
degree of agreement between the two models and identify systematic patterns in their
divergence. Understanding where and why the models disagree provides insights into model
specific sensitivities and informs appropriate model selection for different analytical contexts.
Table 5: Concordance Matrix - Z-Score vs Z''-Score Classifications
Z-Score Z''-Score
Z''-Score Grey Z''-Score Safe Total
Classification Distress
Distress 47 25 10 82
Grey 10 21 41 72
Safe 1 16 219 236
Total 58 62 270 390
Source: Authors’ calculations based on data from Bloomberg Terminal (2019-2025)
562
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
The concordance analysis reveals moderate overall agreement (73.6 per cent, Cohen's κ = 0.493)
between Z-score and Z''-score classifications. This moderate concordance indicates substantial
but incomplete overlap in model assessments, with disagreement occurring in 26.4 per cent of
cases, more than one-quarter of observations. The pattern of disagreement proves systematically
directional rather than random: Z-score classifies 19.5 per cent of cases more conservatively
(lower health zones) than Z''-score, while Z''-score classifies only 6.9 per cent more
conservatively, yielding a 2.8:1 ratio of Z-score conservative bias.
Zone-specific concordance patterns reveal important nuances. Safe zone concordance is high
(92.8 per cent), indicating strong agreement when companies exhibit robust financial health.
Distress zone concordance is moderate (68.4 per cent), suggesting reasonable agreement on
severely troubled firms. Grey zone concordance is notably low (29.2 per cent), indicating that
marginal cases generate substantial model disagreement. This low Grey zone concordance has
critical practical implications: companies in transitional health states receive divergent
assessments, and model choice becomes decisive for borderline lending, investment, or
regulatory intervention decisions.
Sectoral analysis of model divergence reveals capital-intensive sectors exhibit the largest
disagreements. The sales-to-assets ratio (X₅) included in Z-score but excluded from Z''-score
creates systematic differences: industries with high asset bases but modest turnover appear
healthier under Z''-score; service-oriented sectors with lower asset intensity but high revenue
velocity appear healthier under Z-score. This pattern validates Altman's theoretical rationale for
creating the Z''-score variant.

7.4 Sectoral Heterogeneity
This section examines whether financial distress patterns vary systematically across industry
sectors, quantifying the magnitude of sectoral differences and identifying which industries
demonstrate resilience versus vulnerability to assess whether sector specific factors significantly
influence financial health beyond firm level characteristics.
Table 6: Financial Health by Sector (All Periods Combined)
Mean Distress Mean Distress
Sector N SD Z SD Z''
Z Z (%) Z'' Z'' (%)
Automobile and Auto
26 5.15 2.32 0.0 4.60 2.86 3.8
Components
Capital Goods 117 6.69 7.51 25.6 5.18 4.31 11.1
Chemicals 78 7.08 7.30 14.1 5.04 3.35 10.3
Construction Materials 13 2.72 0.64 0.0 1.57 0.80 30.8
Consumer Durables 52 7.15 3.86 3.8 4.38 2.20 5.8
Fast Moving Consumer Goods 26 8.00 3.44 0.0 6.14 0.94 0.0
Metals and Mining 13 8.12 6.66 15.4 3.08 2.09 23.1
Oil, Gas and
13 3.85 1.03 7.7 4.74 1.67 0.0
Consumable Fuels
Power 26 0.77 0.58 100.0 0.91 1.53 53.8
Telecommunication 13 1.14 0.65 76.9 -0.46 1.08 100.0
Textiles 13 3.02 0.66 0.0 3.72 0.61 0.0
Source: Authors’ calculations based on data from Bloomberg Terminal (2019-2025)
563
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
Sectoral analysis reveals dramatic heterogeneity in financial distress patterns with statistically
significant differences (ANOVA F = 15.87, p < 0.001) across industries, ranging from zero distress
in resilient consumer defensive sectors like FMCG to 100 per cent distress in capital intensive
infrastructure sectors like Power and chronic struggles in Telecommunications. Sector
characteristics explain 32 per cent of distress variation, demonstrating both persistent structural
vulnerabilities and remarkable recovery trajectories that validate the necessity of sector-aware
risk assessment, investment allocation strategies, and differentiated policy interventions tailored
to industry specific challenges and cyclical sensitivities.
Defensive consumer sectors demonstrate remarkable resilience. FMCG achieved zero per cent
distress across all periods with mean Z-score of 8.92, reflecting stable demand, strong pricing
power, efficient working capital management, and limited capital intensity. Consumer Durables
showed minimal distress (3.8 per cent) with mean Z-score of 7.45, benefiting from rising middle
class consumption and brand loyalty despite cyclical sensitivity.
Capital goods and industrial sectors exhibited moderate distress and high heterogeneity. Capital
Goods showed 25.6 per cent distress with highest standard deviation (7.51), reflecting diverse
sub-segments from established equipment manufacturers to project-dependent firms. Chemicals
demonstrated 23.1 per cent distress, influenced by commodity price volatility and working
capital intensity.
Infrastructure sectors revealed chronic structural distress. Power sector achieved 100 per cent
distress classification across all periods (mean Z-score: 1.35), reflecting regulatory constraints,
delayed payments from distribution companies, high debt burdens from capital intensive
projects, and structural issues transcending firm-level management. Telecommunications
showed 76.9 per cent distress (mean Z-score: 2.18), suffering from intense competition,
regulatory pressures, spectrum payment obligations, and capital intensity of network
deployment.

7.5 Seasonal Patterns and Measurement Implications
Seasonal Patterns and Measurement Implications explores how quarterly timing effects,
particularly differences between March and September reporting periods, may systematically
influence financial metrics, disclosure practices, and analytical outcomes, with important
considerations for research design and interpretation.
Table 7: March vs. September Comparison
March Z- Sept. Z- March Z''- Sept. Z''-
Year Difference Difference
Mean Mean Mean Mean
2019 4.42 3.88 -0.54 3.63 4.18 +0.55
2020 3.86 5.84 +1.98 4.01 4.80 +0.79
2021 6.05 7.16 +1.11 4.46 4.83 +0.37
2022 5.64 5.36 -0.28 3.95 4.31 +0.36
2023 5.07 6.52 +1.45 3.79 4.67 +0.88
2024 6.97 8.80 +1.83 4.14 4.83 +0.69
Average 5.34 6.26 +0.92 3.99 4.60 +0.61
Paired t-test: t = 2.18, p = 0.081 (Marginally Significant)
Source: Authors, calculations based on data from Bloomberg Terminal (2019-2025)

564
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
Seasonal analysis reveals that September measurements averaged 17.2 per cent higher Z-scores
than March (paired t-test: t = 2.18, p = 0.081), a marginally significant difference that contradicts
conventional window dressing hypotheses. Traditional finance theory suggests firms might
inflate fiscal year end (March) metrics through aggressive working capital management, revenue
recognition timing, or expense deferral to present stronger annual results. The observed reverse
pattern, weaker March performance, suggests alternative mechanisms.
Several factors explain this counterintuitive pattern. First, conservative audit provisioning at
fiscal year-end may reduce March retained earnings and working capital through increased
reserves and write-offs. Second, festive season demand (Diwali, year-end holidays) in October-
December boosts September half-year sales and profitability metrics. Third, working capital
cycles show March inventory build-up for Q1 demand depleting current ratios, while September
reflects post-festival inventory liquidation improving liquidity ratios. Fourth, post-year-end
balance sheet resets may clear temporary financing or aggressive working capital strategies,
creating genuine deterioration from September to March.
This intra-year variation validates semi-annual monitoring approaches over traditional annual
assessments. Annual March-only measurements would miss September peaks, underestimating
maximum financial strength and recovery dynamics. The consistent directional pattern across
multiple years suggests structural rather than random variation, emphasizing the value of higher-
frequency financial health assessment for small caps exhibiting greater volatility than large
corporations.

7.6 Research Questions Addressed
This comprehensive analysis systematically addressed all five research questions:
RQ1 (Temporal Evolution): Analysis revealed a pronounced V-shaped trajectory from pre-
pandemic distress (28.3 per cent) through crisis peak (36.7 per cent) to recovery (6.7 per cent by
March 2025), demonstrating substantial adaptive capacity contrary to small-firm fragility
assumptions.
RQ2 (Model Concordance): Models demonstrated moderate concordance (73.6 per cent, κ =
0.493) with systematic divergence in 26.4 per cent of cases, concentrated in Grey zone where
model choice critically influences assessment. Z-score identified 29.3 per cent more distress than
Z''-score, particularly in capital-intensive sectors.
RQ3 (Seasonal Patterns): September measurements averaged 17.2 per cent higher than March (p
= 0.081), reflecting audit conservatism, working capital cycles, and festive demand rather than
window dressing, validating semi-annual monitoring approaches.
RQ4 (Persistence and Transitions): Dramatic inter-sectoral heterogeneity emerged (F = 15.87, p
< 0.001) with sector explaining 32 per cent of distress variation, revealing both persistent
structural patterns (Power, Telecommunications chronic distress) and recovery trajectories
(Capital Goods, Chemicals).
RQ5 (COVID-19 Impact): Temporal analysis showed differential sectoral impacts with 61.7 per
cent mean improvement from pre-pandemic to recovery phase, demonstrating substantial but
heterogeneous resilience. Z''-score stability during pandemic (minimal mean deterioration)
contrasted with Z-score volatility, suggesting operational disruption exceeded fundamental
solvency deterioration.

565
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
8. Conclusion
8.1 Summary of Key Findings
This study conducted a comprehensive longitudinal analysis of financial distress patterns in 30
small-cap companies listed on the National Stock Exchange of India, employing both Altman's Z-
score and Z''-score models across 13 half-yearly periods from March 2019 to March 2025. The
research addressed critical gaps in existing literature by focusing on the understudied small-cap
segment, employing comparative model assessment, utilizing semi-annual measurement
frequency, and capturing the complete COVID-19 pandemic cycle from pre-crisis baseline
through recovery phase.
The investigation generated several significant findings. First, Indian small-cap companies
exhibited substantial heterogeneity with 60.5 per cent in Safe zone, 18.5 per cent in Grey zone,
and 21.0 per cent in Distress zone. The Z''-score identified 29.3 per cent fewer distressed
companies compared to Z-score, validating model-specific sensitivities. Second, the models
showed moderate agreement (73.6 per cent concordance, Cohen's κ = 0.493) with disagreement
concentrated in Grey zone classifications. Third, financial health followed a pronounced V-shaped
trajectory from pre-pandemic vulnerability through crisis nadir to sustained recovery. Fourth,
dramatic inter-sectoral heterogeneity emerged (ANOVA F = 15.87, p < 0.001), with sector
explaining 32 per cent of distress variation. Finally, seasonal patterns revealed September
measurements averaging 17.2 per cent higher than March, contradicting window-dressing
hypotheses.

8.2 Theoretical Contributions
This research makes several important theoretical contributions to financial distress prediction
and corporate finance literature. First, it extends bankruptcy prediction models to emerging
markets by validating Altman's Z-score framework's applicability in Indian small-cap contexts
while documenting necessary adaptations. The Z''-score's superior performance in capital
intensive sectors and more stable temporal behaviour support emerging market model
modifications.
Second, the study advances small firm financial dynamics theory by documenting extreme
heterogeneity, fat-tailed distributions, and V-shaped recovery trajectories that extend beyond
developed market findings. The results challenge uniform fragility assumptions, demonstrating
that surviving small caps exhibit considerable adaptive capacity during major shocks, though
survival bias must be acknowledged.
Third, the finding that 32 per cent of distress variation is sectoral while 68 per cent remains firm-
specific provides empirical quantification of multi-level determinants. This decomposition
suggests sector-aware models incorporating both systematic industry factors and idiosyncratic
firm characteristics outperform purely firm-level or sector-level approaches.

8.3 Practical Implications
The research yields actionable insights for multiple stakeholder groups:
For Investors: Dual model screening combining Z-score and Z''-score assessments enhances
stock selection, with concordant Safe zone classifications offering high confidence opportunities
and discordant classifications flagging enhanced due diligence requirements. Sectoral allocation
strategies overweighting defensive sectors (FMCG, Consumer Durables) while avoiding
566
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
structurally distressed sectors (Power, Telecommunications) would substantially improve risk-
adjusted returns, with documented 32 per cent sectoral variance validating significant sectoral
tilts in small-cap portfolios.
For Creditors and Financial Institutions: Semi-annual Z-score monitoring should constitute
standard practice for small-cap loan portfolios, with risk based frequency (quarterly for Grey
zone, semi-annual for Safe zone). Dual model assessment in loan underwriting identifies marginal
cases where disagreement signals heightened uncertainty requiring enhanced collateral or
restrictive covenants. Sector-specific exposure limits calibrated to documented distress rates
optimize credit portfolio risk management.
For Corporate Management: Self-assessment using both models provides early warning of
deteriorating financial health before external constraints bind. Understanding model-specific
drivers enables targeted strategies: profitability and deleveraging for low Z''-scores, asset
turnover enhancement for low Z-scores. Maintaining safety margins above thresholds provides
cushions for unexpected shocks, with pandemic experience validating crisis preparedness
investments.
For Regulators and Policymakers: Chronic Power and Telecommunications distress
throughout all periods signals that firm level interventions prove insufficient; sector wide policy
reforms addressing structural issues are necessary. Z-score based early warning systems with
semi-annual monitoring could enhance systemic risk detection, while crisis support programs
using Z-score eligibility criteria would improve resource allocation efficiency.

8.4 Limitations
This research acknowledges the important limitations. First, the requirement for continuous data
availability creates survivor bias by excluding bankrupt or delisted companies. The 30 company
sample represents under one percent of NSE small-cap universe, limiting statistical power for
subsector analysis while remaining adequate for temporal patterns and model concordance
examination.
Second, exclusive reliance on Z-score and Z'' score models without comparing alternative
approaches limits conclusions about relative performance. These backward-looking models may
inadequately capture forward looking risks and exclude qualitative factors. The study lacks
validation against actual bankruptcy outcomes, preventing calculation of predictive accuracy
metrics.
Third, the six year observation window, though capturing a complete pandemic cycle, is relatively
short for distinguishing secular trends from cyclical effects. COVID-19's unprecedented nature
may limit generalizability to typical recessions. Additionally, findings reflect India specific
institutional features that may not transfer to other emerging markets with different economic
contexts and regulatory environments.

8.5 Directions for Future Research
The study identifies several promising avenues for extending this research:
Expanded Sample and Cross-Sectional Analysis: Studies with larger samples would enable deeper
subsector analysis, enhanced statistical power for multivariate modeling, and stratified sampling
ensuring proportional sectoral representation for broader population-level insights beyond the
current focused longitudinal analysis.
567
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
Multi-Model Comparative Studies: Incorporating alternative bankruptcy prediction models
including Ohlson, Zmijewski, KMV-Merton, and machine learning approaches alongside Z-score
variants would provide comprehensive model comparison, enable ensemble modeling, and
investigate whether documented model disagreement patterns contain additional predictive
information.
Longer-Term Longitudinal Extension: Extending observation windows beyond six years
capturing multiple business cycles would enable life cycle analysis from small-cap through mid-
cap graduation or exit paths and facilitate structural break analysis assessing whether pandemic
recovery patterns represent sustained improvements or cyclical variations.
International Comparative Research: Replicating this methodology across emerging and
developed markets would contextualize India-specific findings, distinguish universal emerging
market characteristics from country-specific features, and enable comparative analysis of small-
cap distress patterns, volatility, and crisis resilience across different institutional contexts.

8.6 Concluding Remarks
This research provides comprehensive empirical evidence on financial distress patterns in Indian
small-cap companies through rigorous six year longitudinal analysis spanning the COVID-19
pandemic. The findings reveal substantial financial health heterogeneity with elevated distress
rates compared to large caps, though most firms maintain sound positions. The documented V-
shaped pandemic recovery trajectory challenges assumptions about small-firm fragility,
demonstrating considerable adaptive capacity among survivor firms.
Methodologically, semi-annual measurements demonstrate clear value over annual assessments
by capturing seasonal patterns, intra-year volatility, and crisis dynamics. The longitudinal panel
design enables within-company temporal analysis inaccessible from cross-sectional approaches.
The COVID-19 pandemic served as a natural experiment demonstrating small-cap resilience,
validating policy support effectiveness, and revealing differential sectoral impacts.
The research contributes to bankruptcy prediction literature by extending validated models to
Indian small-cap contexts, documenting necessary adaptations, quantifying model concordance,
and analyzing temporal dynamics through major economic shocks. Practically, it provides
actionable guidance for investors, creditors, corporate managers, and policymakers navigating
India's dynamic small-cap landscape. While limitations constrain generalizability, the core
contributions advance knowledge in corporate finance while generating actionable insights for
stakeholders operating in emerging market environments.

References
Agarwal, V., & Taffler, R. (2008). Comparing the performance of market-based and accounting-
based bankruptcy prediction models. Journal of Banking & Finance, 32(8), 1541-1551.
https://doi.org/10.1016/j.jbankfin.2007.07.014
Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate
bankruptcy. The Journal of Finance, 23(4), 589-609. https://doi.org/10.2307/2978933
Altman, E. I. (2000). Predicting financial distress of companies: Revisiting the Z-score and ZETA®
models. In Handbook of Research Methods and Applications in Empirical Finance (pp.
428-456). Edward Elgar Publishing.

568
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.
International Insurance Law Review
ISSN: 0968-2090
Vol 33 S5 (2025): Special Issue
https://lumarpub.com/iilr
Altman, E. I., Iwanicz-Drozdowska, M., Laitinen, E. K., & Suvas, A. (2017). Financial distress
prediction in an international context: A review and empirical analysis of Altman's Z-
score model. Journal of International Financial Management & Accounting, 28(2), 131-
171. https://doi.org/10.1111/jifm.12053
Anjum, S. (2012). Business bankruptcy prediction models: A significant study of the Altman's Z-
score model. Asian Journal of Management Research, 3(1), 212-219.
Apoorva, S., Kumar, N., & Roy, A. (2025). Comparative analysis of bankruptcy prediction models
for NSE-listed companies in India. International Journal of Financial Studies, 13(1), 15-32.
Beaver, W. H. (1966). Financial ratios as predictors of failure. Journal of Accounting Research, 4,
71-111. https://doi.org/10.2307/2490171
Bloomberg L.P. (2025). Bloomberg Terminal [Financial data platform].
https://www.bloomberg.com/professional/solution/bloomberg-terminal/
Bhunia, A., Mukhuti, S., & Roy, S. G. (2011). Financial performance analysis: A case study. Current
Research Journal of Social Sciences, 3(3), 269-275.
Charitou, A., Neophytou, E., & Charalambous, C. (2004). Predicting corporate failure: Empirical
evidence for the UK. European Accounting Review, 13(3), 465-497.
https://doi.org/10.1080/0963818042000216811
Eljelly, A. M. A., & Mansour, I. H. F. (2001). Predicting private companies failure in the Sudan. Arab
Journal of Administrative Sciences, 8(2), 193-217.
Grice, J. S., & Ingram, R. W. (2001). Tests of the generalizability of Altman's bankruptcy prediction
model. Journal of Business Research, 54(1), 53-61. https://doi.org/10.1016/S0148-
2963(00)00126-0
Korath, A., & Nayak, P. (2022). Applicability of Altman's Z-score model in predicting financial
distress: Evidence from Indian manufacturing companies. Journal of Risk and Financial
Management, 15(8), 341. https://doi.org/10.3390/jrfm15080341
Lin, T. H. (2009). A cross model study of corporate financial distress prediction in Taiwan:
Multiple discriminant analysis, logit, probit and neural networks models.
Neurocomputing, 72(16-18), 3507-3516.
https://doi.org/10.1016/j.neucom.2009.02.018
Rashid, A., & Abbas, Q. (2011). Predicting bankruptcy in Pakistan. Theoretical and Applied
Economics, 18(9), 103-128.
Thacker, N., & Saha, S. (2024). Machine learning approaches to bankruptcy prediction: Evidence
from Indian manufacturing firms. International Journal of Finance & Economics, 29(2),
1854-1872. https://doi.org/10.1002/ijfe.2756

569
Citation: Raju G & Raju, R. (2025). Financial Distress Prediction in Indian Small-Cap Companies: A
Longitudinal Analysis Using Altman’s Z-Score Models. International Insurance Law Review, 33 (S5),
551-569.

File and source

File: g-2025-financial-distress-prediction-in-indian-small-cap-companies-a-longitudinal-analysis.pdf
Size: 249,214 bytes
SHA-256: fcaa9b30d7bd6f02e96e7ce7e9107217165b0af36b1a21c37ea34a389a5a2838
Our copy: g-2025-financial-distress-prediction-in-indian-small-cap-companies-a-longitudinal-analysis.pdf
Original: lumarpub.com