Prediction of firm’s financial crash-Evidence from the Factor

Analysis Model-

Nesrin Benhayoun, Ikram Chairi, Amina El Gonnouni, Mariam Tanana, Abdelouahid

Lyhyaoui
Laboratory of Innovatives Technologies, National School of Applied Sciences, Abdelmalek Essaadi
University, Tangier, Morocco

ABSTRACT: This paper seeks to examine the importance of financial indicators in the finan-
cial health of companies and how we can use these relevant attributes to predict financial
crashes, including financial risks that have significantly increased in the context of the current
global financial crisis.
Given that a company forms a micro financial system, its good financial health contributes to
building a powerful economic engine; therefore, we have selected a sample of 20 firms over 3
years (2009-2011), and we have extracted about forty indicators that impact on the firms’ fi-
nancial health, in order to set up our Factor Analysis Model through which we can predict
firms’ creditworthiness and default risk.

Keywords: Firm; Financial crashes; Financial health; Default risk; Factor Analysis

1 INTRODUCTION

The global financial crisis of 2008 makes the management of financial risk a strategic regula-
tory factor, not only for bank structures but also for any financial or economic entity, regardless
of its size and the sector in which it undertakes its activity.
Although many regulatory regimes have established their rules (Basel II, Basel III…) of fi-
nancial risk control from the consequences of different financial crises and from their impact on
the solvency of institutions, it is by no means easy to study and measure this key, and so there is
still a long way to go in this area.
In fact, financial market regulations come in the wake of the financial crises which have
been recurring phenomena in the history of economy; thus, it is relevant to cite some landmark
financial crashes that have occurred over the last two hundred years:
 1974: Herstatt Bank (German) crisis caused a huge crisis in foreign exchange liquidity,
and it caused several more bank failures, mainly the banks to which Herstatt Bank owed
the delivery of foreign currency. The Herstatt crisis is well known in international fi-
nance as “Herstatt risk,” and it had many implications for the regulatory framework.
 1994: Metallgesellschaft, a German conglomerate, revealed publicly that its “Energy
Group” was responsible for losses amounted to $1.5 billion, due mainly to cash-flow
problems resulting from large oil forward [Link] Metallgesellschaft crisis had
many implications for the authorities to explore how proper supervision could have
averted disaster and how similar financial crises may be avoided in the future
 1998: the failure of Long-Term Capital Management (LTCM) is said to have nearly
blown up the world’s financial system. Indeed the fund’s woes threatened to create ma-
jor losses for its Wall Street lenders. LTCM was so big that the Federal Reserve Bank
of New York took the unprecedented step to facilitate a bailout of the private hedge
fund, out of fear that a forced liquidation might ravage world markets.
  2008: the global financial crisis or global economic crisis is commonly believed to have
begun in July 2007 with the credit crunch, when a loss of confidence by US investors in
the value of sub-prime mortgages caused a liquidity crisis. By September 2008, the cri-
sis had worsened as stock markets around the globe crashed and became highly volatile.
The housing collapse in the US is commonly referred to as the trigger of the global fi-
nancial crisis, especially with a series of banks and insurance companies failures such
as the collapse of Lehman Brothers in September 2008.
It is believed that the financial system needed better regulation and required unprece-
dented government intervention.
Although it is very difficult to define the notion of risk, risk is related to a negative occur-
rence that is caused by external or internal vulnerabilities that we cannot predict.
Theoretically, we can distinguish these different risks that a financial institution can meet: De-
fault risk, Downgrading risk, Interest rate risk, Currency risk, Operational risk, Liquidity risk
and Risk strategy.
Thus, when it comes to measuring financial risk in its entirety, the question is more compli-
cated due to the lack of enough of historical data as required for analysis. Since companies and
banks form the micro financial system, we can focus our studies on the parameters affecting mi-
cro financial health with the purpose to develop our Factor Analysis Model as an instrument for
predicting the firm’s default risk and for assisting the financial institution in decision-making
through the default risk prediction.
This paper is structured as follows. Section 2 presents an overview of the default risk. Sec-
tion 3 shows detailed description of the financial indicators which reflect the financial health of
the company. Section 4 reviews how this relevant attributes can be used to asses and predict the
firm’s default risk using the Factor Analysis Model. Finally, the last section provides some con-
cluding comments.

2 DEFAULT RISK

Default risk is the uncertainty surrounding a firm’s ability to service its debts and obliga-
tions. Prior to default, there is no way to discriminate unambiguously between firms that will
default and those that will not. At best, we can only make probabilistic assessments of the like-
lihood of default.
Although these risks do not seem large, they are in fact highly significant. First, they can in-
crease quickly and with little warning. Second, the margins in corporate lending are very tight,
and even small miscalculation can undermine the profitability of lending.
But most importantly, many lenders are themselves borrowers, with the high level of lever-
age. Unexpected realizations of default risk have destabilized, de-capitalized and destroyed
lenders. Banks, finance institutions, and insurers: none have escaped unscathed.
Default risk cannot be hedged away, or structured away. The government cannot insure it
away. It is a reflection of the substantial risk in companies’ futures. Various schemes exist, and
more are coming, which can shift risk, but in the end, someone must bear this risk. It does not
“net out” in the aggregate.
The risk of firm’s default affects virtually every financial contract. Therefore the pricing of
default risk has received much attention; both from lender who have to ensure its claims and
from traders who have a strong interest in pricing transactions accurately. For this purpose, it is
extremely important to be able to know about the future firm’s financial behavior and to assess
the degree of firm’s default.

3 FINANCIAL RATIOS ANALYSIS

The aim of this study is to investigate the use of financial ratios to ascertain state of financial
health of firms and to assist the financial institution in decision-making.
Through this study, we demonstrate the use of actual financial data for financial ratio analy-
sis in order to show exactly how a financial institution can predict how well the financial be-
havior of one company will perform relative to that of another and how we can overcome the
difficulties in applying the principles of financial ratio analysis when the data are not homoge-
neous as is the case in our samples.
The financial analysis is the selection, evaluation, and interpretation of financial data to as-
sist in investment and financial decision-making. Financial analysis may be used internally to
evaluate issues such as the efficiency of operations, and credit policies, and externally to evalu-
ate potential investments and the creditworthiness of borrowers.
The primary source to draw the financial data needed in financial analysis is the data provid-
ed by the company itself in its annual report and required disclosures. The annual report com-
prises the income statement, the balance sheet, and the statement of cash flows, as well as foot-
notes to these statements.
In this reading, we have selected, analyzed and interpreted the pertinent ratios, enabling
judgments on the current and future financial condition and operating performance of the com-
pany, as provided below.
A ratio is a mathematical relation between one quantity and another. A financial ratio is a
comparison between one bit of financial information and another.
Ratios can be classified according to their general characteristics; in this study we have cho-
sen the classification as follow:
The Operating ratios: When we assess a company’s operating performance, we want to know
if it is applying its assets in an efficient and profitable manner. The operating ratios show the
efficiency of a company’s management and used in expense control, and in measuring the prof-
itability and financial soundness of a firm. We have selected:
 Working Capital ratio: Indicates whether a company has enough short term assets to
cover its short term debt. It measure of both a company’s efficiency and its short
term financial health.
 Working Capital Requirement: Indicates the minimum amount of resources that a
company requires to effectively cover the usual costs and expenses necessary to op-
erate the business.
 Cash ratio: It is commonly used as a measure of company liquidity. It can therefore
determine if, and how quickly, the company can repay its short term debt.
 Cash flow ratio: It is the cash resulting from income generating activities minus ex-
penses and investments. It is a key metric for any entity that handles cash, and it can
determine if the company can finance its operations through the cash it generates
from ongoing activities.
The Financial ratios: When we assess a company’s financial condition, we want to know if it
is able to meet its financial obligations. It can be used to analyze trends and to compare the
firm’s financials to those of other firms. We have selected:
 Productivity ratio: It represents the efficiency with which physical inputs are con-
verted to useful outputs. It can be computed by the ratio between the value added
and the turnover. This is an indicator of company’s ability to harness physical and
human resources to maximize its production of goods and services.
 Leverage ratio: It measures how much money a company should safely be able to
borrow over long periods of time. It does this by comparing the company’s total debt
and dividing it by the amount of owner’s equity.
 Coverage ratio: A measure of company’s ability to meet its financial obligations, it
compares a company’s total debt to its cash flow and it provides an indication of
company’s ability to cover total debt with its yearly cash flow from operations.
 Solvency ratio: It provides an assessment of the likelihood of a company to continue
congregating its debt obligations. It compares a company’s owners equity to its total
assets. Generally speaking, the lower a company’s solvency ratio, the greater the
probability that the company will default on its debt obligations.
The Profitability ratios: Compare components of income with sales. They give us an idea of
what makes up a company’s income and are usually expressed as a portion of each dollar of
sales. We distinguish:
 Operating Profit Margin: This is a ratio that indicates how much of each dollar of
sales is left over after operating expenses and it compares the operating income of
company to its turnover.
  Net Profit margin: This is a ratio of net income to turnover, and indicates how much
of each dollar of sales is left over after all expenses.
There are hundreds of ratios that can be formed using available financial statement data. The
ratios selected for our study depend on the type of our analysis about the credit worthiness and
the type of our sample of firms. We will see in the next section how to use these ratios to get
prediction of firm’s default risk through a Factor Analysis Model.

4 FACTOR ANALYSIS MODEL

Many Scientific studies are featured by the huge number of variables used. Because of these
big numbers of variables that are into play, the study can become rather complicated. For situa-
tion such as these, factor analysis has been invented. The goal of factor analysis is to reduce
“the dimensionality of the original space and to give an interpretation to the new space, spanned
by a reduced number of new dimensions which are supposed to underlie the old ones” (Rietveld
& Van Hout 1993: 254).
The aim of this section is to show the use of our sample selection to build the factor analysis
model allowing the measure of the firm’s financial health.

4.1 Theoretical Model

The Factor Analysis Model is one of several techniques which seek to explain the correlation
between a set of variables by a smaller set of random variables. It uses a small number of imag-
inary variables to express the basic data structure by studying the internal relationship between
the variables, and reflects the main information and interdependence of these original data.
The following assumes that the p observed variables (The Xi) that have been measured for each
of the n subjects have been standardized.

X1= a11F1+…a1mFm+e1
X2= a21F1+…a2mFm+e2
.
.
.
Xp = ap1F1+…apmFm+ep

The Fj are the m common factors, the ei are the p specific errors, and the aij are the factor pxm
factor loadings.
In matrix form this can be written as:

Xpx1= Apxm Fmx1 + epx1

It doesn’t make sense to use factor analysis if the different variables are unrelated; this is why
the starting point of factor analysis is a correlation matrix as provided below.

4.2 Correlation Matrix and Principal Factor Analysis (PFA) approach

The correlation Matrix provides the intercorrelations between the studied variables. The di-
mensionality of this matrix can be reduced by looking for variables that correlate highly with a
group of other variables, but correlate very badly with variables outside of that group. These
variables with high intercorrelations could well measure one underlying variable, which is
called a factor.
This research investigates financial data over 3 years 2009-2011, which all come from 20
Moroccan companies that belong to different sectors and have different sizes.
Firstly, we have selected 39 variables, over 3 years, which are impacting the financial health
of companies:
Table 1. Selected variables.
V1 Turnover V14 Equity / Total assets V27 Financial costs / Gross op-
erating profit
V2 Net equity V15 Working capital / Working V28 Financial costs / Operating
capital requirement cash surplus
V3 Net cash V16 Leverage ratio V29 Gross operating profit /
Turnover
V4 Net profit V17 Coverage ratio V30 Net profit margin ratio
V5 Working capital V18 Change in debts / Cash flow V31 Net profit / Net equity
V6 Working capital require- V19 Rotation net cash V32 Net profit / Permanent capi-
ment tal
V7 Value added V20 Inventory turnover V33 Long term debts / Cash flow
V8 Gross operating margin V21 Delay of payment of cus- V34 Net profit / Equity
tomers
V9 Gross operating profit V22 Delay of payment to ven- V35 Staff costs
dors
V10 Operating cash surplus V23 Gross margin / Turnover V36 Lenders
V11 Free cash flow V24 Cash flow / Turnover V37 State
V12 Cash flow V25 Productivity ratio V38 Current operating income
V13 Solvency V26 Staff costs / Value added V39 Non-operating income

When we assess a firm’s default risk, we want to know if a company is solvent, if it is profit-
able and if it is still productive; this is why we have computed the correlation coefficients, us-
ing SPSS 10, between our 39 variables involved and the three relevant components (threshold =
0.5): Solvency, productivity and profitability, in order to detect the variables that influence the
most. Through the correlation analysis of 2009 and 2010 as provided in the appendix (figures 1
to 6), we can construct the vector prediction for 2011 from the selected variables, and through
the calculation of the correlation coefficients and according to this approach, we obtained three
optimal vectors of default risk prediction of companies, devoted to measure the solvency risk,
the productivity risk and the profitability risk.
Such as mentioned in the appendix (figures 1 to 6), the three vectors are composed as follows

Table 2. Vector Analysis.

Solvency Vector Productivity Vector Profitability Vector
Solvency ratio Productivity ratio Net profit margin ratio
Equity / Total assets Change in debts / Cash flow Solvency
Rotation net cash Gross margin / Turnover Leverage ratio
Inventory Turnover Net profit / Equity Inventory turnover
Gross margin / Turnover Delay of payment of customers
Financial costs / Gross operating Net profit / Net equity
profit
Net profit margin ratio Net profit / Permanent capital

These vectors determine the substantive importance of particular variables to compute the
risk of the three relevant vectors and predict on average the default risk through the scores of
each component.
For a more thorough precision of the risk prediction, the dimensionality of the correlation
matrix can be reduced by the use of Principal Factor Analysis (PFA) approach using SPSS10,
in order to obtain factors that create a new dimension and can be visualized as classification ax-
es along which measurement variables can be plotted.
As can be seen, the matrix in the appendix (table 3) that presents 8 factors as the number of
factors to be retained, instead of 39 variables. The eight factors based on the data from 2009-
2010 can explain 86% of the variance contribution, which means the model has a good measur-
ing effect.
So, we can reduce the matrix dimension to eight factors to predict the default risk.
5 CONCLUSION AND SUGGESTION

This study has selected a sample of financial data of 20 companies over 2009-2011 to meas-
ure the default risk through the solvency risk, the productivity risk and the profitability risk, us-
ing the Factor Analysis Model that provided eight principal factors that can be used as a classi-
fication axes to forecast the default risk of a company.
In order to ensure the availability and the accuracy of data, this research discarded a number
of indicators which have difficulties in acquisition or differences in the method of calculation;
for example, besides information that companies are required to disclose through financial
statements, other information will have a certain degree of precision constraints of measuring
risk and need to further improve, such as the market prices of securities of publicly-traded cor-
poration and information on stock price indices for industries and for the market.
Therefore, more methods can be used and more results can be integrated to get the more ex-
act risk prediction. Such methods as Neural Network and SVM model that should be used to
improve the accuracy of measuring and forecasting.

REFERENCES

Basel Committee on Banking Supervision, Basel III International framework for liquidity risk measure-
ment, standard and monitoring (December 2010), Bank for International Settlements: 3-38
Allen, F., A. Babus, and E. Carletti (2009), Financial Crises: Theory and Evidence, Annual review of Fi-
nancial Economics 1, forthcoming.
Basel Committee on Banking Supervision, Working Paper No. 13, Bank Failures in Mature Economies
(April 2004), Bank for International Settlements: 1-5.
Culp, Christopher and Miller, Merton, Risk Management Lessons from Metallgesellschaft, Journal of Ap-
plied Corporate Finance, 1994.
Kolman, Joe, 1999, LTCM Speaks, Derivatives Strategy (April): 12-17
Pacelle, Mitchell, Randall Smith, and Anita Raghavan, 1999, Investors May See LTCM, the Sequel, Wall
Steet Journal (May 20): C1.
Crosbie.P, Bohn.J, Modeling Default Risk, Modeling Methodology, 2003 Moody’s KMV Company: 10-
30.
Duffie, D., and K.J. Singleton, 1995, Modeling Term Structures of Defaultable Bonds, working paper,
Stanford Graduate School of Business.
Grahman, John R. and Campbell R Harvey. The Theory and Practice of Corporate Finance: Evidence
from the Field, Journal of Financial Economics,60 2001:187-243.
Bartholomew D. (1984): The Foundation of Factor Analysis, Biometrika 71: 221-32.
Sharma, S. (1996). Applied Multivariate Techniques. United States, John Wiley & Sons.
Field, A. (2000), Discovering Statistics using SPSS for Windows, London – Thousand Oaks –New Delhi:
Sage publications.
Richard O. Duda, Peter E. Hart, David G. Stork, Pattern Classification, Second Edition: 215-330.
APPENDIX

Figure [Link] Correlation result for 2009.

Figure [Link] Correlation result for 2010.

Figure [Link] Correlation result for 2009.

Figure [Link] Correlation result for 2010.

Figure [Link] Correlation result for 2009.

Figure [Link] Correlation result for 2010.

Table 3. Principal Factor Analysis.
Variables Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6 Factor 7 Factor 8
V1 0,921 -0,082 0,077 0,042 -0,021 -0,050 -0,109 -0,061
V2 0,995 0,020 -0,022 -0,049 0,013 -0,008 -0,032 -0,015
V3 -0,299 0,072 -0,095 0,747 0,401 0,081 0,166 0,188
V4 0,986 -0,006 -0,023 0,073 0,061 -0,003 -0,004 0,001
V5 0,941 -0,005 -0,023 -0,135 -0,009 -0,005 -0,006 -0,012
V6 0,934 -0,016 -0,007 -0,242 -0,070 -0,017 -0,031 -0,040
V7 0,996 -0,031 -0,004 0,008 0,025 -0,013 -0,019 -0,013
V8 0,997 -0,032 0,000 -0,007 0,016 -0,015 -0,025 -0,017
V9 0,990 -0,031 -0,009 0,035 0,039 -0,010 -0,010 -0,009
V10 0,912 -0,035 -0,024 0,335 0,178 0,014 0,040 0,054
V11 0,873 -0,031 -0,030 0,404 0,201 0,026 0,063 0,077
V12 0,901 0,009 -0,045 0,362 0,211 0,019 0,045 0,070
V14 0,299 0,571 0,010 0,016 -0,076 0,194 0,263 -0,192
V15 -0,010 -0,014 0,000 0,144 -0,309 -0,179 -0,219 0,586
V16 -0,070 -0,320 -0,170 -0,266 0,521 0,488 -0,175 -0,086
V17 0,068 0,313 0,899 -0,105 0,176 0,062 0,002 0,088
V18 0,041 0,381 0,849 -0,001 0,101 -0,013 0,098 0,039
V19 -0,026 0,469 -0,066 0,351 -0,204 0,405 -0,290 -0,143
V20 -0,090 -0,091 -0,076 -0,220 0,163 -0,065 0,762 0,210
V21 0,215 0,226 -0,345 -0,067 0,139 -0,510 0,159 0,058
V22 -0,032 0,585 -0,400 -0,142 0,040 0,051 -0,133 0,273
V23 0,318 0,432 -0,384 -0,219 -0,154 0,320 0,403 0,256
V24 0,026 0,878 -0,282 -0,064 0,141 0,028 -0,150 0,001
V26 0,974 -0,030 0,016 -0,098 -0,026 -0,023 -0,056 -0,026
V27 0,920 -0,006 0,010 -0,209 -0,075 -0,021 -0,077 -0,036
V28 0,987 -0,029 -0,012 0,083 0,064 -0,009 -0,003 -0,001
V29 0,991 -0,028 -0,012 0,032 0,040 -0,009 -0,006 -0,007
V30 -0,911 0,040 -0,014 0,345 0,135 0,044 0,064 0,084
V31 -0,045 0,007 0,038 0,133 -0,020 -0,422 -0,142 -0,302
V32 -0,031 -0,736 -0,059 0,093 -0,061 0,172 -0,092 0,021
V33 0,036 -0,159 0,084 -0,130 -0,054 0,220 -0,365 0,580
V35 0,537 -0,341 0,077 -0,005 -0,309 0,351 0,242 0,175
V36 0,060 0,871 -0,277 -0,066 0,089 0,011 -0,175 -0,017
V37 0,209 0,249 0,209 0,329 -0,697 -0,159 0,115 0,066
V38 -0,024 0,085 0,023 0,165 -0,346 0,431 0,161 -0,371
V39 0,068 0,313 0,899 -0,105 0,176 0,062 0,002 0,088