American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
Influence of Firm Financial Characteristics on Leverage of Manufacturing Firms Listed
Companies in Nairobi Securities Exchanges
Author Details: Helen Wairimu
Department of Accounting and Finance-School of Business, University of Nairobi
Abstract
Securities exchanges have greater roles to play in regard to economic and social development in both
developed and developing economies. Hence, this paper examined the influence of firm financial characteristics
on the leverage of manufacturing listed companies in Nairobi securities exchange. Panel data was collected
from annual financial statements and panel regression modelling was adopted to analyze the data. Operating
cash flows had a significant moderating effect on the influence of firm financial characteristics on the leverage
of manufacturing listed firms in Nairobi securities exchange.
Key Words: Financial characteristics, Firm size, Profitability, Tangibility, Growth opportunities
Introduction
Securities exchanges have greater roles to play in regard to economic and social development in both developed
and developing economies (Padaya, 2016). They are supposed to act as a medium through which both deficit
and surplus financial units are able to raise finances to fund their growth opportunities, provide currency
market, facilitate public and private investment and provide debt funding platform (Mwangi, 2016).
Although African bond and equity markets are still underdeveloped as compared to European, American, Asian
and Australian securities markets, there is need to improve on liquidity which is hindering the development
(Association of Securities Exchange in Africa, ASEA, 2014). Despite the hurdles facing securities markets,
there are recorded changes in combined value of total equities and bonds traded from US$454 974.4 million in
equities and US$ 2080.6 billion in bonds in 2013 to $325.0 billion in equities, $1.2 trillion in bonds in 2015,
and $438.0 billion in electronic transfer finds and others, representing a market capitalization of over $1.3
trillion (ASEA, 2015), This has improved capital access within developing economies (ASEA, 2015).
Leverage refers to the proportion of debt to equity in the capital structure of a firm. There are two types of
leverage; financial leverage which is defined as the use of debt financing by the firms and operational leverage.
Following Harc (2015) the operational definition for purposes of measuring leverage in this study is calculated
as the ratio of long-term debt to total assets and total debt to total assets. Indeed, (Ezeoha, 2008; Mwangi, 2016)
used a similar definition. Bandyopadhyay & Barua (2016) used similar measures in India.
Long term debt is a portion of debt financed in more than one accounting cycle and short term is paid back
within a single accounting period (Mwangi, 2016). Long term debt is also referred to as non-current liabilities
and is at times preferred by firms since it gives them time to make profits to indemnify it or pay immediate
expenses like research and development for start-up businesses. A firm which is highly indebted, whether by
short or long term, is likely to suffer distress.
Moreover, empirical findings (Strebulaev & Yang, 2006; Shubita & Alsawalhah, 2012) show that firms’
exposure to financial risk is linked to their inability to service loans as per their contractual agreement. If this is
prolonged, the firm could eventually be faced with financial distress, erosion of the equity and subsequently
winding up (Madan, 2007). Consequently, the current study sought to:
i.
ii.
iii.
To determine the influence of tangibility of assets on the leverage of manufacturing firms listed at
Nairobi Securities Exchange.
To find out the influence of growth opportunities on the leverage of manufacturing firms listed at
Nairobi Securities Exchange.
To establish the influence of firm size on the leverage of manufacturing firms listed at Nairobi Securities
Exchange.
http://www.abrj.org
Page 139
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
iv.
To examine the influence of profitability on the leverage of manufacturing firms listed at Nairobi
Securities Exchange.
v. To evaluate the moderating effect of operating cash flows on the influence of financial characteristics on
the leverage of manufacturing firms listed at Nairobi Securities Exchange.
Literature Review
Pecking order theory was brought forth by Myers (1984) and it stipulates that there is always a financing pattern
which is followed commencing from internal financing, debt financing and final issue of external equity.
According to Donaldson (1961), internal equity is more preferred because an organization always wishes to
minimize flotation costs which are associated with external financing. The preference for external finance rather
than issue new equity is based on the fact that issue of new debt attracts lower flotation costs compared to the
later (Myers, 1984; Myers & Majluf, 1984).
A UK case which drew 3500 unquoted small and medium enterprises (SMEs) by Hall et al. (2000) revealed
inverse and significant relationship between profitability and short-term debt financing. Moreover, the age of
the firm had an influence on leverage decision whereby older and young firms had a negative influence on
financing decision. In a subsequent study by Hall et al. (2004), it was asserted that leverage decision is
dependent on firm’s ability to generate more revenue, therefore, those which were generating more they had
lower chances of borrowing. According to Myers (2001) those companies which have the potential of making
huge revenue they will rely more on internally generated resources to finance their financial needs.
A comparative analysis between static trade-off theory and pecking order model by Shyam-Sunder and Myers
(1999) argued that the pecking order theory has a superior influence on firms financing decision. Further, Frank
and Goyal (2003) reported a positive and significant relationship between tangible security and leverage
decision since assets can be used as collateral security. Moreover, an Australian case which considered SMEs
revealed superiority of pecking order financing as compared to static trade off theory.
Although some empirical inquiries had supported the superiority of pecking order model, Fama and French
(2002) supported an inverse relationship between profitability and leverage but disclaimed the findings because;
increased profitability can signal investment opportunities and there are chances of increased fixed cost. Indeed,
whenever a firm generates more revenue it is easier to offset the debt. In contrast, a study by Fama and French
(2005) revealed that most of the firms which were listed in 1973-2002 violated the applicability of pecking
order financing model and opted for equity financing. In fact, Frank and Goyal (2003) proved that in America it
is not possible for listed companies to fully satisfy their financing needs and they opt for debt financing to meet
financing speed.
There is a preference for equity financing against debt financing since the level of information asymmetry
associated with debt financing is higher as compared to equity financing (Fama & French, 2005). Indeed, firms
have recently opted for employees to share ownership schemes and right issues. This is to minimize the
possibility of ownership structure changes. Mwangi (2016) argued that there are low chances of breaching
information grip while issuing new shares or rights issues as compared to debt financing which may attract
binding covenant. The theory is appropriate for the study since the study seeks to examine the moderating of
operating cash flow on the influence of firm characteristics on leverage decision. From the foregoing literature
review, the following relationship was conceptualized.
http://www.abrj.org
Page 140
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
Tangibility of Assets
Non-current assets/Total assets
Profitability
Return on Assets
Firm size
Sales volume/Total assets
Leverage
Short term debt to total assets
Long term debt to total assets
Total debt to total assets
Growth Opportunities
Market value
Book value
Operating Cash Flows
Operating
Cashflows/Total assets
Methodology
The study used panel data. Panel data is a series of multidimensional data where the behaviour of entities are
observed over time (Wooldridge, 2002). The key advantage of panel data is the ability to allow the researcher to
control for variables that are not observable or measurable like culture and management practices over time but
not across entities (Wooldridge, 2002). It was obtained from the NSE handbooks and from specific companies’
websites. As shown in the data collection sheet data on non-current assets, market prices, book value, turnover,
total liabilities, profit after tax, operating cash flows were gathered. Secondary data was collected for period
2008-2016. Univariate and multivariate techniques were applied for data analysis. The influence was tested
through the use of regression analysis and moderation was examined through an examination of marginal
changes of slope coefficients due to the introduction of operating cash flows. The general models were of the
form:
Lit=𝛽0 + 𝛽1Ti,t+ 𝛽2 Si,t+ 𝛽3 Gi,t+ 𝛽4Pi,t+ 𝓔j……………………………..……………………Model 1
The following regression model with the moderating variable was used for the analysis as proposed by Baron
and Kenny (1986).
Lit=𝛽0 + 𝛽1Ti,t+ 𝛽2 Si,t+ 𝛽3 Gi,t+ 𝛽4Pi,t + 𝛽5 CFi,t + CFi,t(𝛽6Ti,t+ 𝛽7Si,t+ 𝛽8 Gi,t+ 𝛽9Pi,t)+
Where
𝓔 j……….Model
2
Lit – Short term liabilities to total assets, long term liabilities to total assets, total liabilities/total assets for each
firm i at time t
T= Tangibility of assets, G=Growth opportunities, S=Firm size, P= Profitability, CF=Operating cash flows, 𝛽i
(i=0,1,2,…9) are the associated regression coefficients, 𝓔j is the associated error term.
Findings and Discussions
Descriptive Statistics for Manufacturing and Allied
As shown in Table 1, the average tangibility amongst listed manufacturing companies in NSE was 0.55, with a
minimum of 0.27 and a maximum of 0.93. Most companies in manufacturing and the allied sector had a high
portion of non-current assets and 45% in current assets. There were minimal variations in asset tangibility as
indicated by the standard deviation of 0.15. A coefficient of skewness of 0.16 revealed that most companies had
high proportions in non-current assets. This implies that they may have enough collateral security to access debt
http://www.abrj.org
Page 141
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
capital. The profitability of manufacturing companies was 14%, with a minimum of -50% and a maximum of
105%. Most companies were profitable with the period under consideration as accounted for by coefficient of
skewness of 1.86. This implies that there may be minimal reliance on debt capital amongst manufacturing if
they were to rely on internally generated finances.
The average growth opportunity amongst manufacturing firms was 1.78, with a minimum of 0.00 and maximum
of 8.13. There were wide variations in growth opportunities as indicated by the standard deviation of 2.
Moreover, most companies were positively skewed as accounted for by 1.53. These findings contrasted the
market timing theory. The average operating cash flows to total assets was 0.21, with a minimum of -2.01 and
maximum of 1.31. These findings were negatively skewed as accounted by the skewness coefficient of -1.47.
There is a need for manufacturing companies to evaluate their working capital operating cycle so as to optimize
benefits associated with prudent working capital management.
There was a high dependency on long term debt as accounted for by an average of 0.18 and a maximum of 1.13.
The positive coefficient of skewness of 2.59 revealed that most firms highly financed their assets using long
term debt. High dependency on long term debt financing can be attributed to the availability of collateral
security. The average short-term debt to total assets was 0.28, with a maximum of 0.67 and a minimum of 0.03.
Negative skewness of -0.03 revealed that had a low reliance on short term debt finance and those who relied had
a minimal variation on its application as indicated by the standard deviation of 0.14. This implies that most
manufacturing companies listed in NSE had adopted conservative working capital management. The average
reliance of total debt to total assets was 0.46, with a minimum of 0.13 and a maximum of 1.45. From the
findings, it can be deduced that some firms had borrowed debts which exceeded their assets requirements this
would pose a threat to their business operations, especially in situations when they needed to borrow more
capital whose access would be curtailed by lack of collateral security.
Table 1 Manufacturing and Allied Sector Descriptive Statistics
N
Minimum
Maximum
Mean
Std. Deviation
Skewness
T
56
0.27
0.93
0.55
0.15
0.16
0.32
P
56
-0.50
1.05
0.14
0.22
1.86
0.32
S
56
12.87
17.98
15.94
1.62
-0.46
0.32
G
56
0.00
8.13
1.78
2.00
1.53
0.32
CF
56
-2.01
1.31
0.21
0.52
-1.47
0.32
LTA
56
0.00
1.13
0.18
0.20
2.59
0.32
STA
56
0.03
0.67
0.28
0.14
-0.03
0.32
DTA
56
0.13
1.45
0.46
0.27
1.28
0.32
Panel Diagnostic Tests
Autocorrelation Test for Manufacturing and Allied Companies Listed in Nairobi Securities Exchange
As shown in Table 2, models with LTA as the response variable had F statistics of 13.41, without cash flow
moderation, and 9.169 with moderation. The p values for both were less than 0.05. The test statistics were
therefore significant in all cases at a 5% level of significance to indicate the presence of first order serial
correlation in the data. The model without moderation where STA is the response variable had an F statistic of
67.275 with a p value of 0.0002 and model with moderation had an F statistic of 9.569 and p value of 0.0213 to
indicate significance at 5% significance level. This implied presence of first order serial correlation. For the
DTA response variable models, the F statistics were 63.325 and 102.48 with p values of 0.0002 and 0.0001
without and with moderation respectively. This, therefore, implies the presence of serial correlation. With the
presence of the first order, serial correlation FGLS models were fitted.
http://www.abrj.org
Page 142
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
Table 2 Woodridge Test for Manufacturing and Allied Companies Listed in Nairobi Securities Exchange
Dependent variable
Model
LTA
Without moderator
With moderator
STA
Without moderator
With moderator
DTA
Without moderator
With moderator
F (1,6)
P value
13.41
0.0106
9.169
0.0232
67.275
0.0002
9.569
0.0213
63.325
0.0002
102.48
0.0001
Multicollinearity Test Statistics for Manufacturing Listed Companies in Nairobi Securities Exchange
Table 3 presents the VIFs for the various study variables. The results indicate that the VIFs were not greater
than 5, hence there was no collinearity amongst independent variables.
Table 3 Multicollinearity Test Statistics for Manufacturing Listed Companies in Nairobi Securities
Exchanges
Variable
CF
S
T
G
P
Mean VIF
VIF
1/VIF
2.14
0.466374
1.88
0.532819
1.52
0.656788
1.25
0.799893
1.11
0.898716
1.58
Heteroskedasticity Test Results for Manufacturing and Allied Companies Listed in Nairobi Securities
Exchange
Table 4 shows the likelihood ratio tests statistics for manufacturing and allied companies listed in NSE. The null
hypotheses of the tests were that the error variance was homoscedastic for each model. The likelihood-ratio
tests produced chi-square values of 46.27, 30.54 and 26.17 with a p-value less than 0.05. This implies that the
test was significant at 5% level of significance hence the existence of heteroscedasticity in the study. To remedy
the problem, FGLS estimation technique was used (Wooldridge, 2002).
Table 4 Heteroskedasticty Test Results for Manufacturing and Allied Companies Listed in Nairobi
Securities Exchange
Response Variable’s models
Chi Square
Degree of freedom
P value
STA
46.27
5
0.000
LTA
30.54
5
0.000
DTA
26.17
5
0.0001
Stationarity Test Results for Manufacturing and Allied Companies Listed in Nairobi Securities Exchange
The unit root test statistics for companies listed in the manufacturing and allied sector in NSE are presented in
Table 5. From the table, it is evident that all variables are stationary at the level since the null hypothesis that all
variables are not stationary at 5% significant level is rejected. This is further assurance on the robustness of the
expected results. Further on, there was no need to differentiate the data.
Table 5 Stationarity Test Results for Manufacturing and Allied Companies Listed in Nairobi Securities
Exchange
http://www.abrj.org
Page 143
American Based Research Journal
Variable
T
P
G
CF
S
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
Statistic
Value
p-value
Inverse chi-squared (84)
P
14.0252
0.000
Inverse normal
Z
5.3247
0.000
Inverse logit t (199)
L*
0.3231
0.000
Modified inv. chi-squared
Pm
0.0048
0.000
Inverse chi-squared (84)
P
33.3902
0.000
Inverse normal
Z
-4.6418
0.000
Inverse logit t (199)
L*
-1.8078
0.000
Modified inv. chi-squared
Pm
3.6644
0.000
Inverse chi-squared (84)
P
86.8161
0.000
Inverse normal
Z
-4.2291
0.000
Inverse logit t (194)
L*
-9.1568
0.000
Modified inv. chi-squared
Pm
13.761
0.000
Inverse chi-squared (78)
P
6.8753
0.000
Inverse normal
Z
3.9246
0.000
Inverse logit t (199)
L*
0.896
0.000
Modified inv. chi-squared
Pm
-1.3464
0.000
Inverse chi-squared (78)
P
10.7145
0.000
Inverse normal
Z
3.1801
0.000
Inverse logit t (199)
L*
0.146
0.000
Modified inv. chi-squared
Pm
-0.6209
0.000
Hausman Test Results for Manufacturing and Allied Companies Listed in Nairobi Securities Exchange
As shown in Table 6 there was enough evidence to warrant rejection of the null hypothesis at 5% level of
significance for LTA model with moderation, STA models with and without moderation and DTA model with
moderation as accounted for by p value of 0.000, 0.0211, 0.0023 and 0.000. Consequently, the appropriate
models to fit were fixed effects regression model. Further, there was not enough evidence to warrant rejection
of the null hypothesis at 5% for LTA and models without moderation since their p values were greater than 0.05
as accounted for by p values of 0.7596 and 0.0855 respectively. Thus, the most appropriate model to fit was the
random effects.
Table 6 Hausman Test Results for Manufacturing and Allied Companies Listed in Nairobi Securities
Exchange
Dependent variable
Model
Chi Square
df
P value
LTA
Without moderator
1.87
4
0.7596
With moderator
31.65
6
0.000
Without moderator
11.54
4
0.0211
With moderator
20.44
6
0.0023
Without moderator
8.17
4
0.0855
With moderator
31.76
6
0.000
STA
DTA
http://www.abrj.org
Page 144
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
Granger Causality Test Results for Manufacturing and Allied Companies Listed in Nairobi Securities
Exchange
As shown in Table 7, the p-values for all lagged financial characteristics (in isolation) values and DTA, run
against DTA, are greater than 5% level of significance. This implies that the null hypotheses that individual
financial characteristic does not granger cause leverage is not rejected for manufacturing and allied listed
companies in NSE. When all lagged values of financial characteristics and DTA were run against DTA at the
same time, the p value was zero. Being less than 5% level of significance, it means that the null hypothesis that
financial characteristics do not granger causes leverage is rejected. It means that the financial characteristics of
a firm, as a combination but not in isolation, can explain its leverage.
When the lagged values of DTA and individual financial characteristic were run against individual financial
characteristics values at the same time, the p value for T and G were less than 5% level of significance. The p
values for S and P were greater than the said significance level.
Table 7 Granger Causality Test Results for Manufacturing and Allied Companies Listed in Nairobi
Securities Exchange
Dependent
Independent (Lagged)
F Statistic
P value
DTA
S, D, T, A
6.06
0.0063
T, D, T, A
4.81
0.0042
P, D, T, A
3.23
0.0149
G, D, T, A
2.58
0.0293
S, T, P, G, D, T, A
38.02
0.000
S
D, T, A, S
0.9
0.9172
T
D, T, A, T
0.9
0.4146
P
D, T, A, P
1.56
0.2249
G
D, T, A, G
2.03
0.1467
FGLS Regression Results of STA as Dependent Variable with Moderator for Manufacturing and Allied
Listed Companies in Nairobi Securities Exchange
As shown in Table 8, results on the effect of financial characteristics on short term debt financing for energy
and petroleum listed companies in NSE while operating cash flow was incorporated in the model show that the
coefficient of SCF was -0.077 hence firm sizes had a negative impact on short term debt financing when the
operating cash flow was incorporated. The p value was 0.000 which is less than 5% level of significance. This
shows that the moderating influence of operating cash flow on firm size was statistically significant on short
term debt financing. The coefficient of TCF was -0.039 hence tangibility had a negative influence on short term
debt as operating cash flow increased. The p value was 0.822 which is less than 5% level of significance. This
indicates that the moderating influence of operating cash flow on tangibility was statistically insignificant on
long-debt financing.
The coefficients of PCF and GCF were -0.173 and -0.06 respectively. This indicates that profitability and
growth opportunities had a positive influence on short debt respectively when operating cash flow was
incorporated. The p values were 0.173 and 0.000 respectively to imply that the moderating influence of
operating cash flow on profitability and growth opportunities were insignificant and significant respectively on
short debt financing at 5% level of significance.
To further confirm the influence of the moderator, the coefficients of the model without the moderator are
compared with the average marginal effect or change of financial characteristics on short term debt financing. If
the two are different then there is moderation else no moderation. The marginal change shows how much shortterm debt changes by with an increase in one unit of the relevant financial characteristic when the average
http://www.abrj.org
Page 145
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
moderator value is incorporated. This is achieved by differentiating model 2 in chapter three partially and
incorporating the average moderating value as follows
STAit
= β1+ β6CF = -0.386-0.039*.21= -0.467
Tit
STAit
= β2+ β7CF =0.0178-0.074*.21= 0.0015
Sit
STAit
= β3+ β8CF =0.054-0.173*.21= 0.018
Pit
STAit
= β4+ β9CF = 0.006-0.0598*.21= -0.0067
Git
Comparison between moderated and non-moderated variables with the operating cash flow revealed that it had
a moderating influence on the influence of firm financial characteristics on short term leverage of
manufacturing and allied companies listed at NSE.
Table 8 FGLS Regression Results of STA as Dependent Variable with and without Moderator for
Manufacturing and Allied Listed Companies in Nairobi Securities Exchange
Without Moderation
With Moderation
Variable
Coefficient
Std. Error
Z
p>z
Coefficient
Std. Error
Z
p>z
cons
-.643
.134
-4.80
0.000
.240
.109
2.21
.027
T
-.075
.102
-.74
.461
-.386
.092
-4.17
.000
S
.061
.007
8.25
.000
.018
.006
2.77
.006
P
.011
.056
.20
.845
.0540
0.029
1.88
0.06
G
-.004
.006
-.67
.502
.006
.003
1.75
.08
CF
.995
.224
4.45
.000
TCF
-.038
.172
-.22
.822
SCF
-.078
.018
-4.33
0.000
PCF
-.173
.127
-1.39
.173
GCF
-.060
.013
-4.44
.000
Wald chi2 (4)
=86.88
R2 = 0.5150
P >
0.00
Chi2
Wald
chi2
=634.5
(9)
R2
0.9211
=
P > Chi2
.0000
FGLS Regression Results of LTA as Dependent Variable with and without Moderator in Manufacturing and
Allied Firms Listed in Nairobi Securities Exchange
As shown in Table 9, results on the effect of financial characteristics on short term debt financing for construction and
allied listed companies in NSE while operating cash flow was incorporated in the model show that the coefficient of SCF
was 0.089 hence firm size had a positive influence on long term debt financing when the operating cash flow was
incorporated. The p value was 0.42 which is greater than the 5% level of significance. This shows that the moderating
influence of operating cash flow on firm size was statistically insignificant on long term debt financing. The coefficient of
TCF was -1.808 hence tangibility had a negative influence on long term debt as operating cash flow decreased. The p
value was 0.013 which is less than 5% level of significance. This indicates that the moderating influence of operating cash
flow on tangibility was statistically significant on long-debt financing.
http://www.abrj.org
Page 146
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
The coefficients of PCF and GCF were -0.943 and -0.078 respectively. This indicates that profitability and growth
opportunities had a negative influence on long term debt respectively when operating cash flow was incorporated. The p
values were 0.055 and 0.217 respectively to imply that the moderating influence of operating cash flow on profitability
and growth opportunities were negative and insignificant respectively on long debt financing at 5% level of significance.
To further confirm the influence of the moderator, the coefficients of the model without the moderator are compared with
the average marginal effect or change of financial characteristics on long term debt financing. If the two are different then
there is moderation else no moderation. The marginal change shows how much long-term debt changes by with an
increase in one unit of the relevant financial characteristic when the average moderator value is incorporated. This is
achieved by differentiating model 2 in chapter three partially and incorporating the average moderating value as follows
STAit
= β1+ β6CF = 0.863-1.808*0.21=0.484
Tit
STAit
= β2+ β7CF = 0.089+0.087*0.21=0.107
Sit
STAit
= β3+ β8CF =-0.035-0.944*0.21=-0.23
Pit
STAit
= β4+ β9CF = -0.014-0.078*0.21=-0.030
Git
Comparison between moderated and non-moderated variables with the operating cash flow revealed that it had a
moderating influence on the influence of firm financial characteristics on long term leverage of listed manufacturing and
allied firms in NSE.
Table 9 FGLS Regression Results of LTA as Dependent Variable with and without Moderator in
Manufacturing and Allied Firms Listed in Nairobi Securities Exchange
Without Moderation
With Moderation
Variable
Coefficient
Std. Error
Z
p>z
Coefficient
Std. Error
Z
p>z
cons
-.740
.133
-5.55
.000
-1.749
.521
-3.36
.001
T
.310
.096
3.23
.001
.863
.305
2.83
.005
S
.042
.008
5.03
.000
.089
.029
3.1
.002
P
.146
.098
1.50
.133
-.035
.137
-.26
.798
G
.010
.010
1
.328
-.014
.014
-1
.327
CF
.228
1.119
.2
.839
TCF
-1.808
.730
-2.48
.013
SCF
.087
.107
.81
.42
PCF
-.943
.491
-1.92
.055
GCF
-.078
0.06
-1.24
.217
Wald chi2
=46.45
(4)
R2 = 0.3256
P >
0.00
Chi2
Wald
30.16
chi2(9)
=
R2 = 0.5233
p>Chi2
.0004
FGLS Regression Results with DTA as dependent Variable with and without Moderator in
Manufacturing and Allied Companies Listed in Nairobi Securities Exchange
As shown in Table 10, results on the effect of financial characteristics on debt financing for construction and
allied listed companies in NSE while operating cash flow was incorporated in the model show that the
http://www.abrj.org
Page 147
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
coefficient of SCF was -0.036 hence firm sizes had a negative influence on debt financing when the operating
cash flow was incorporated. The p value was 0.737 which is greater than the 5% level of significance. This
shows that the moderating influence of operating cash flow on firm size was statistically insignificant on debt
financing. The coefficient of TCF was -1.519 hence tangibility had a positive influence on term debt as
operating cash flow increased. The p value was 0.034 which is less than 5% level of significance. This indicates
that the moderating influence of operating cash flow on tangibility was statistically significant in debt financing.
The coefficients of PCF and GCF were 1.037 and -0.089 respectively. This indicates that profitability and
growth opportunities had a positive influence on long term debt respectively when operating cash flow was
incorporated. The p values were 0.031 and 0.158 respectively to imply that the moderating influence of
operating cash flow on profitability and growth opportunities were significant and insignificant respectively on
debt financing at 5% level of significance.
To further confirm the influence of the moderator, the coefficients of the model without the moderator are
compared with the average marginal effect or change of financial characteristics on debt financing. If the two
are different then there is moderation else no moderation. The marginal change shows how much debt changes
by with an increase in one unit of the relevant financial characteristic when the average moderator value is
incorporated. This is achieved by differentiating model 2 in chapter three partially and incorporating the average
moderating value as follows
STAit
= β1+ β6CF = 0.444-1.519*0.18 =0.170
Tit
STAit
= β2+ β7CF =0.125-0.036*0.18 =0.118
Sit
STAit
= β3+ β8CF =0.0291-1.037*0.18=-0.158
Pit
STAit
= β4+ β9CF = -0.014 -0.089*0.18 =-0.027
Git
Comparison between moderated and non-moderated variables with the operating cash flow revealed that it had
a moderating influence on the firm financial characteristics on the leverage of listed manufacturing and allied
firms in Nairobi Securities exchange.
http://www.abrj.org
Page 148
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
Table 10 FGLS Regression Results of DTA as Dependent Variable with and without Moderator in
Manufacturing and Allied Companies Listed in Nairobi Securities Exchange
Without Moderation
With Moderation
Variable
Coefficient
Std. Error
Z
p>z
Coefficient
Std. Error
Z
p>z
cons
-1.286
.163
-7.92
.000
-1.756
.494
-3.56
0.000
T
.133
.102
1.30
.193
.444
.345
1.29
.198
S
.103
.009
11.05
0.00
.125
.025
5.03
0.000
P
-.093
.114
-.81
.417
0.029
0.139
0.21
0.834
G
-.002
.012
-.14
.885
-.014
0.014
1
.327
CF
1.625
1.145
1.42
.156
TCF
-1.519
.718
-2.12
.034
SCF
-0.036
.108
-.34
.737
PCF
-1.037
.482
-2.15
.031
0.06
-1.24
.217
GCF
-.078
Wald chi
=137.60
2
(4)
2
R = 0.5917
P >
0.00
Chi
2
Wald
98.86
2
chi (9)
=
2
R = 0.7184
p>Chi2
.0000
Conclusion and Recommendations
Based on the findings manufacturing companies listed in Nairobi should evaluate their leverage policy and
adopt a short term management strategy that would match their business operational capacity. They should
examine their borrowing capacity based on asset tangibility, growth opportunities, profitability, and firm size.
Adherence to pecking order while seeking financial of listed companies will not only protect asset tangibility of
listed companies but also minimize boost investors’ confidence since they have more control over their
investment. Management and professional bodies ought to develop manuals and financial simulation models
which are geared towards educating and sensitizing management of listed companies on the most viable
financing alternative.
References
i.
ASEA (2013). Association of African Securities Exchanges Annual Report.
ii.
ASEA (2014). Association of African Securities Exchanges Annual Report.
iii.
ASEA (2015). Association of African Securities Exchanges Annual Report.
iv.
Bandyopadhyay, A., & Barua, M. (2016). Factors Determining Capital Structure and Corporate
Performance in India: Studying the Business Cycle Effects. The Quarterly Review of Economics and
Finance, 6(1): 160-172.
v.
Baron, R. M., & Kenny, D. A. (1986). The Moderator–Mediator Variable Distinction in Social
Psychological Research: Conceptual, Strategic, and Statistical Considerations. Journal of personality
and social psychology, 51(6),1173-1182.
vi.
Donaldson G. (1961). Corporate debt capacity: A study of corporate debt policy and the determination
of corporate debt capacity, Boston, division of Research, Harvard Graduate School of Business
Administration.
http://www.abrj.org
Page 149
American Based Research Journal
Vol-8-Issue-3 March-2019 ISSN (2304-7151)
vii.
Ezeoha, A. E. (2008). Firm size and corporate financial leverage choice in a developing economy:
Evidence from Nigeria, The Journal of Risk Finance, 9(4), 351-364.
viii.
Fama E.F., and French K.R. (2005). Financing Decisions: Who Issues Stock? Journal of Financial
Economics, 7(6), 549-582.
ix.
Fama, E. F., & French, K. R. (2002). Testing Trade-off and Pecking Order Predictions about Dividends
and Debt. Review of financial studies, 15(1), 1-33.
x.
Frank, M. Z. & Goyal, V. K. (2003). Testing the Pecking Order Theory of Capital Structure. Journal of
Financial Economics, 67(2),217-228.
xi.
Hall G., Patrick, J., Hutchinson & Michaelas, N. (2000). Industry Effects on the Determinants of
Unquoted SMEs’ Capital Structure. International Journal of the Economics of Business, 7(3), 297 –
312.
xii.
Hall, C., Patrick, J., Hutchinson & Michaelas, N. (2004). Determinants of the Capital Structures of
European SMEs. Journal of Business Finance & Accounting. 31(5), 711-728.
xiii.
Harc, M. (2015). The Relationship between Tangible Assets and Capital structure of Small and
Medium-Sized Companies in Croatia, Econviews, 28(1), 213–224.
xiv.
Madan, K., (2007). An Analysis of The Debt‐Equity Structure of Leading Hotel Chains in India.
International Journal of Contemporary Hospitality Management, 19(5), 397-414.
xv.
Mwangi, J. M. (2016). Effect of Financial Structure on Financial Performance of Firms Listed at East
Africa Securities Exchanges, (PHD Thesis), Jomo Kenyatta University of Agriculture and Technology,
School of Business, Nairobi, Kenya.
xvi.
Myers S. C., & Majluf, N. S. (1984). Corporate Financing and Investment Decisions when Firms have
Information that Investors do not have. Journal of Financial Economics,13(2),187-221.
xvii.
Myers, S. C. (1984). The Capital Structure Puzzle. Journal of Finance, 39(3), 575-592.
xviii.
Myers, S. C. (2001). Capital Structure. Journal of Economic Perspectives. 15(2), 81-102.
xix.
Padaya, B. (2016). Impact of Financial Leverage on Market Value Added: Empirical Evidence from
India. Journal of Entrepreneurship, Business and Economics, 4(2), 40-58.
xx.
Shubita, M. F., & Alsawalhah, F. (2012). The Effect of Capital Structure and Profitability. International
Journal of Business and Social Science, 3(16),104-112.
xxi.
Shyam-S, L; & Myers, S. C., (1999). Testing Static Trade-Off Against Pecking Order Models of Capital
Structure. Journal of Financial Economics, 51(2),219–244.
xxii.
Strebulaev, I. A., & Yang, B. (2006). The Mystery of Zero-Leverage Firms. Accessed on
http://dx.doi.org/10.2139/ssrn.890719 on 15/10/2017.
xxiii.
Wooldridge, J. M. (2002). Econometric Analysis of Cross Section and Panel Data. Cambridge, MA:
MIT Press.
http://www.abrj.org
Page 150