Journal of Finance and Economics 103
a variant of NPV. If you know the annuity of an
investment, and how many years it should generate net
cash-inflows or outflows, then you can easily calculate its
NPV by discounting the annuity with the relevant
weighted average cost of capital.
3. Data and Descriptive Statistics
To analyse what determines the choice of capital
budgeting methods in Sweden, a questionnaire (Appendix
1) was sent in 2005 and 2008 to the CFOs of all Swedish
companies listed on the Stockholm Stock Exchange. To
facilitate a comparison between the surveys, the questions
were the same as used by Graham and Harvey (2001) and
Brounen et al. (2004). [4] If no executive had the title
CFO, then the questionnaire was sent to another senior
executive (controller, treasurer, or CEO) responsible for
financial management.
In 2005, the questionnaire was sent to 244 companies
by postal mail three times, with response deadlines 8
January, 14 March, and 23 May. Non-respondents by the
first deadline were contacted by phone to encourage them
to respond. In the end, 112 questionnaires (45.9%) were
returned. However, seven were not useable and were
dropped, leaving an adjusted response rate of 43.0%.
In 2008, the questionnaire was sent to 249 companies
by postal mail four times, with response deadlines 18
February, 10 March, 3 April, and 16 June. Again, non-
respondents by the first deadline were contacted by phone.
In the end, 92 (36.9%) questionnaires were returned. Four
were not useable, leaving an adjusted response rate of
35.3%. The total adjusted response rate for the two
surveys was thus 39.1%, compared to 9% for Graham and
Harvey (2001) and 5% for Brounen et al. (2004).
In simple probit analysis of the response rate the
probability of response was statistically significantly
higher for larger companies, and the probability to answer
the survey was higher in 2005 compared to 2008. Industry
classification, P/E-ratio, degree of leverage, and dividend
pay-out level did not have statistically significant effects
on the probability of response (Table A1 in Appendix 2).
The questionnaire made clear that questions regarding
capital investment referred to all non-routine capital
investments accepted or rejected at group/parent-company
level. The reason for this framing was that otherwise, i.e.,
if questions were taken to refer to all investments in the
company (including investments accepted or rejected at
subsidiary level), then the respondents probably would not
be able to give credible answers.
3.1. Dependent Variables
The questionnaire (Appendix 1) consists of three main
questions regarding the use of capital budgeting and cost-
of-capital estimation techniques. [5] Data from question
one, “How often do you use the following capital
budgeting methods (on a scale of 0 to 4, with 0 = never
and 4 = always)?”, was used to construct our dependent
variables. The capital budgeting methods asked about
were:
a). Net-present value (NPV)
b). Internal rate of return (IRR)
c). Annuity
d). Earnings multiple
e). The adjusted present value (APV)
f). Pay-back
g). Discounted pay-back
h). Profitability index
i). Accounting rate of return (ARR)
j). Sensitivity analysis
k). Value-at-risk (VAR)
l). Real options
The number and share of respondents reporting that
they occasionally or never use or frequently or always use
each method are reported in Table 2. Annuity, adjusted
present value (APV), value-at-risk, and real options were
far less used than the other methods. We excluded them
from further analysis as not providing sufficient variation
to analyze.
Table 2. Number and proportion of companies that used each capital
budgeting Method never or occasionally (0-2) vs frequently or
always (3-4)
Capital budgeting method 0-2 % 3-4 %
(a) NPV 75 38.86% 118 61.14%
(b) IRR 135 69.95% 58 30.05%
(c) Annuity 187 96.89% 6 3.11%
(d) Earnings multiple approach 139 72.02% 54 27.98%
(e) APV 180 93.26% 13 6.74%
(f) Pay-back 88 45.60% 105 54.40%
(g) Discounted pay-back 160 82.90% 33 17.10%
(h) Profitability index 169 87.56% 24 12.44%
(i) ARR 147 76.17% 46 23.83%
(j) Sensitivity analysis 106 54.92% 87 45.08%
(k) VAR 180 93.26% 13 6.74%
(l) Real options 189 97.93% 4 2,07%
The recommended methods used frequently or always
by the most listed companies in Sweden were the net
present value (61%) and sensitivity analysis (45%), but
not recommended pay-back method was used frequently
or always by 54% of the respondents.
Mean values, standard deviations, and differences in
mean values for the most used capital budgeting methods
in 2005 and 2008 are reported in Table 3. Higher mean
values indicate more extensive use of the method.
Table 3. Mean values, standard deviations, and differences in mean
values for the most used capital budgeting methods in 2005 and 2008
2005 2008
Dependent variable Mean Sd Mean Sd Difference
(a) NPV 2.50 1.38 2.55 1.38 0.05
(b) IRR 1.57 1.58 1.27 1.45 -0.30***
(d) Earnings multiple 1.36 1.61 1.41 1.57 0.05
(f) Pay-back 2.39 1.41 2.20 1.47 -0.19
(g) Discounted pay-back 1.08 1.42 0.74 1.31 -0.34***
(h) Profitability index 0.69 1.26 0.72 1.20 0.03
(i) ARR 1.14 1.55 1.05 1.45 -0.10
(j) Sensitivity analysis 1.92 1.63 2.05 1.53 0.12
Note: *** indicates that the difference is statistically significant at the
1% level.
The differences between 2005 and 2008 are small,
usually not significantly different from zero. The biggest
differences are that IRR and the discounted pay-back

+13062052269

info@desklib.com

What determines the use of capital budgeting methods?

Contribute Materials

Related Documents