- Email: [email protected]
fundamental sensitivities under financial constraints
Accepted Manuscript Title: Fixed Investment/Fundamental Sensitivities under Financial Constraints Author: Nihal Bayraktar PII: DOI: Reference:
S0148-6195(14)00031-9 http://dx.doi.org/doi:10.1016/j.jeconbus.2014.05.001 JEB 5683
To appear in:
Journal of Economics and Business
Received date: Revised date: Accepted date:
10-10-2013 28-5-2014 30-5-2014
Please cite this article as: Bayraktar, N.,Fixed Investment/Fundamental Sensitivities under Financial Constraints, Journal of Economics and Business (2014), http://dx.doi.org/10.1016/j.jeconbus.2014.05.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ip t
us
cr
Fixed Investment/Fundamental Sensitivities under Financial Constraints
Nihal Bayraktar Pennsylvania State University - Harrisburg
te
d
M
an
While most models with financial market imperfections predict investment by financially constrained firms to be more sensitive to financial variables, contracting models argue that investment by such firms should be more sensitive to fundamental determinants of investment because fundamentals capture both investment opportunities and changes in the financial position. By first grouping U.S. manufacturing firms as either financially constrained or unconstrained, this paper examines systematic differences in investment/fundamental sensitivities. The findings show that, as expected of contracting models, investment by financially constrained firms is more responsive to fundamentals. These fundamentals are captured by two prominent empirical measures: profitability shocks and mandated investment rate.
Ac ce p
Keywords: financial market imperfections, contracting models, fixed investment, fundamentals, profitability shocks JEL Classification Number: D2, E22, G1, G3
1 Page 1 of 60
1
Introduction
The investment literature identifies fundamental determinants of investment as one of the main
ip t
determinants of fixed capital investment at the firm level. However, there are different views on investment/fundamental sensitivities of, especially, financially constrained firms. For example,
cr
many models in the financial market imperfections literature expect investment by financially
us
constrained firms to be less sensitive to fundamentals but more responsive to financial variables. Conversely, according to contracting models, which study costly borrowing of firms in the
an
presence of financial frictions, financially constrained firms’ investment behavior is predicted to be more responsive to fundamentals when compared to the investment behavior of financially
M
unconstrained firms. This is due to the fact that changes in fundamentals capture changes in not only investment opportunities but also financial positions. If the shock to fundamentals is
d
persistent, it improves the expected marginal benefit of investment. However, for the same
te
reason, it also improves the terms of trade in financial transactions of the firm. If the expected
Ac ce p
profit increases, the probability of default declines with the increasing value of the firm. Therefore, the agency cost, which compensates risk neutral lenders for expected loss from default, must also decrease.1
This paper, in light of such arguments in the literature, tries to understand whether or not
1
This negative relationship between the profitability shock (one measure of fundamentals) and the agency cost is
firmly established by the literature on defaultable debt, for instance, Carlstrom and Fuerst (1997), Bernanke et al. (2000), Cooley and Quadrini (2001), and Hennessey and Whited (2007) for corporate finance, Chatterjee et al. (2007) for consumer finance, Marcet and Marimon (1992) and Cooley et al. (2004) for long term contract. In the literature, it is clear that default history and current earnings are the most important factors in determining credit limits and interest rates for any unsecured debt financing in reality.
2 Page 2 of 60
investment by expected-to-be financially constrained firms really responds more strongly to shocks to fundamentals than investment by financially unconstrained firms. When compared to previous empirical studies, one main contribution of the paper is the use of alternative measures
ip t
of fundamentals. In the literature, there are many concerns about the predictive power of one of the most commonly used fundamental determinants of fixed capital investment: Tobin’s q (the
cr
ratio of asset market value of a firm to its replacement cost of capital). Alternatively, empirical
us
papers based on investment models with non-convex adjustment costs introduce new measures of fundamentals. These present a forward-looking behavior – an important feature that can allow
an
fundamentals to better capture investment opportunities. In this paper two of these are used as fundamental determinants of investment: profitability shocks and the mandated investment rate
d
included for the purpose of comparison.
M
(a gap measure between the desired and actual capital stocks).2 In the analyses, Tobin's q is also
te
The analyses in the paper are based on a reduced form investment equation, in which both fundamental determinants of investment and financial variables are taken as explanatory
Ac ce p
variables, as well as their interaction terms in some specifications. The regressions are separately run for financially constrained and unconstrained firms. Based on the regression results, systematic differences in the estimated coefficients of the fundamental variables of financially constrained and unconstrained firms are investigated. A firm-level panel data set is constructed from the COMPUSTAT database. The set includes U.S. manufacturing firms for the period of 1983-1996. Different firm characteristics are used to identify financially constrained firms. The
2
“Fundamental Q” would be another good candidate (Gilchrist and Himmelberg, 1995 and 1998; Del Boca et al.,
2008). It has not been included in the study because it has been already reported in the literature that the significance of financial variables drops and fundamentals get more significant with "Fundamental Q."
3 Page 3 of 60
criteria are the level of capital stock, number of employees, dividend to capital ratio, dividend payout, debt-to-capital ratio, firms' bond rating, and the KZ (Kaplan and Zingales) index. The ratio of cash flow to capital, sales to capital, and working capital to capital are the financial
ip t
variables included in the analyses.
The empirical findings support the prediction of contracting models. Firms with financial
cr
constraints exhibit a stronger investment-fundamental sensitivity when compared to the group of
us
firms that are less constrained financially. Even though it is not the main purpose of the paper, in the regression outcomes we can also observe investment/financial variable sensitivities across
an
firm classifications because the regression specifications already include financial variables in addition to fundamentals. In the analysis, it can be seen that the investment/financial variable
M
sensitivity of financially constrained firms is lower in many cases than the investment/financial
d
variable sensitivity observed in the group of unconstrained firms.
te
The rest of the paper is organized as follows. Section 2 presents a contracting model of investment. Section 3 gives information about the link between investment, fundamentals, and
Ac ce p
financial variables, and highlights some empirical issues. Section 4 presents some details on the data set and variables, including the construction of the fundamentals. In Section 5, the empirical results are presented. Section 6 concludes.
2 A Contracting Model of Investment One of the main models of investment, the Q theory, presents a formal link between a firm's investment and marginal q. Furthermore, it shows that marginal q should be the sole determinant of investment. However, the explanatory power of Tobin's q (empirical marginal q) in the investment process has been found to be negligible in many empirical studies; and thus the literature started searching for new investment models. 4 Page 4 of 60
The introduction of financial market imperfections in investment models is one such examples. The financial market imperfections literature assumes that firms' net worth determines their financial position. On the one hand, a firm with low net worth can be considered financially
ip t
constrained, since it is likely to face an asymmetric information problem in financial markets, which makes it difficult for them to find enough external funds to finance their investment
cr
projects. Because of this, their investment decisions are expected to be highly correlated with
us
their internal funds, but not fundamentals. Firms with high net worth, on the other hand, are expected to have a smaller asymmetric information problem; thus, they can find enough external
an
funds to finance their capital adjustment and follow the investment process suggested by changes in fundamentals.
M
Contracting models, a type of financial market imperfections model, also study costly
d
borrowing of firms in the presence of financial frictions. Improving fundamentals, such as
te
profitability of firms, signal not only improvements in investment opportunities but also improvements in firms’ financial position through easier borrowing conditions and declining
Ac ce p
agency costs. This in turn leads to an even higher level of investment. In light of this argument, investment of financially constrained firms is expected to be more sensitive to fundamentals. The following model introduces financial frictions through contracts. The aim is to determine why improving fundamentals may lead to healthier financial position of firms. The model combines the costly state verification framework by Townsend (1978) and Cooley and Quadrini (2001) with the firm-level investment model introduced by Bayraktar (2009). The model creates financial frictions in the market through the costly state verification. The firm honors debt obligations only when the value of taking this action is greater than the value of defaulting. In the
5 Page 5 of 60
occasion of default, the investor verifies the firm's revenue, which is costly but eliminates the incentive problem. The lender, who lends R.Bt+1 today, gets Bt+1 tomorrow if the firm does not default, where R
ip t
is the discounted price of the bond less than one. It is taken as a function of expected profit shocks, capital, and debt. If the firm defaults, the lender renegotiates the debt burden of the firm
cr
in a way to make the firm indifferent with regard to continuing the project or exiting the market.
an
us
The value function, V(⋅), is defined as:
subject to
. In the function, t denotes time period,
(1) (⋅) is the present
M
discounted future value of the firm, C(⋅) is the investment cost function, It stands for investment, Kt is the current capital stock, δ is the depreciation rate,
(⋅) is the profit function, β is the fixed
d
discount factor, and At is the profitability shock. Since profitability shocks are highly auto-
(⋅) are homogenous of degree one in investment and capital, and
Ac ce p
assumed that both C(⋅) and
te
correlated, the expected value of future profitability shock (EAt+1) given At can be calculated. It is
C(⋅) is a convex function. Let
denote the renegotiated debt, which becomes effective if the firm defaults. The
renegotiated debt burden must satisfy the condition of the firm’s indifference between continuing the project and exiting the market: 0 = V(EAt+1|At, Kt+1,
). The reservation utility is assumed
to be zero without loss of generality. The condition implicitly defines the renegotiated debt as a function of At+1 and Kt+1. Due to the possibility of default, the discounted price of the bond, R(⋅), must satisfy the following zero profit condition to convince a risk neutral lender to hold the risky debt:
6 Page 6 of 60
(2) where prob is the probability of no-default corresponding to the cases where the value of At+1 is > 0 (i.e. no-default set).
is the verification cost
ip t
high enough to produce
cost. Given
and >0, in this case,
.
is an increasing function of expected
us
It is straightforward to show that
cr
per unit of debt issue. In case of default, the lender receives renegotiated debt minus verification
profitability shocks given At (i.e.
, which is the main fundamental measure of
an
investment in the empirical part of this paper. In other words, the agency cost, because, for any given set of
M
⁻¹ ≤ β⁻¹, is a decreasing function of
.
d
choices made by the firm (Kt+1, Bt+1), the probability of default is decreasing in
te
3 Investment and Fundamentals: Empirical Issues
Ac ce p
The contracting model presented in the previous section indicates that the profitability shock is a shock to both fundamentals and to finance. Essentially, this means that the shock to the profit cannot be classified solely as a fundamental factor, or other way of saying, as a factor that does affect only the profit but not the finance. This is important because this possibility has not been considered in many empirical studies in the literature. For example, one of the common structures used in the cash-flow/investment sensitivity literature is a regression equation given by
inv = bx + cFV + u,
(3)
where inv is the investment rate. While x is a fundamental measure used to capture investment opportunities (mostly Tobin’s q), and FV stands for financial variables such as net worth or 7 Page 7 of 60
internal funds. u is the error term. This regression equation estimates the sensitivity of investment to changes in internal funds, FV, controlling for investment opportunities, x. In the literature, the equation is estimated separately for financially constrained and unconstrained
ip t
firms which are identified based on a priori proxies.
The capital imperfections literature, on the one hand, has shown that the sensitivity of
cr
investment to financial variables is higher for financially constrained firms, and FV is significant
us
but x is not. On the other hand, another group of researchers have found that finance does not matter, since x is significant but FV is not. While the second group criticizes the first group by
an
arguing that the source is mismeasurement of x, the first group criticizes the second group by arguing that the source is mismeasurement of FV. This seems to be an almost endless debate in
M
the investment literature. Even when a researcher observes x without any measurement error, the
d
researcher may not be able to escape this endless debate.
te
Let us consider two different firms to highlight the view point of the contracting models: one firm financially constrained and the other financially unconstrained. For the financially
Ac ce p
unconstrained firm, a good shock today affects the investment decision only through its effect on the expected marginal efficiency of capital tomorrow. However, for the financially constrained firm, the good shock affects the investment decision not only through the effect on the expected marginal efficiency of capital but also through the effect on the agency cost of borrowing today. In this case, financial frictions, if exist, provide an amplification effect through the agency cost channel. In this regard, two predictions can be made. First, if a researcher correctly identifies the shock and regresses investment on the shock variable, x, the coefficient must be significant regardless of whether or not the firm is financially constrained, assuming the shock is persistent.
8 Page 8 of 60
This presents another interpretation issue: is the coefficient significantly positive because a good shock improves the forecast of fundamentals, decreases the agency cost of borrowing or both? Second, if one firm is financially constrained but the other is not, the coefficient of the
ip t
fundamental shock variable might be greater for the financially constrained firm due to the additional amplification channel of the agency cost of contracting models.3
cr
At this point, we need to remember that if the profitability shock is correctly identified, its
us
coefficient in an empirical investment equation must convey information about both the fundamentals and the finance variable. Therefore, one cannot infer that the finance does not
an
matter since x is significant but FV is not (or less significant than x), because the effect of financial frictions must show up in the coefficient for x, among others, if there are financial
M
frictions. Thus, in order to study how financial frictions affect the investment decision, it is
d
necessary to check whether or not there are systematic differences in the coefficient of x among
te
firms with different characteristics instead of only testing the significances of x versus FV. In this regard, it is expected that firms with similar profitability shocks (x) can produce a different level
Ac ce p
of investment due to disparities in their financial positions. In light of this discussion, the paper analyzes systematic differences between financially constrained and unconstrained firms in their investment sensitivity to fundamentals, including profitability shocks. Another source of disagreement in the literature is how to appropriately measure fundamentals. Despite the availability of a large range of financial variables to determine firms' financial position, most empirical studies use only Tobin's q as a proxy for investment 3
The predictions made from a development in financial frictions literature, for example, Cooley and Quadrini
(2001) in the framework of short-term standard debt contract, and Clementi and Hopenhayn (2006), Albuquerque and Hopenhayn (2004) and Cooley et al. (2004) in the framework of long-term contract show that investment of financially constrained firms responds more to fundamentals or profitability shocks.
9 Page 9 of 60
opportunities. Theoretically, the correct measure to capture investment opportunities is marginal q, which is defined as the present discounted value of future profits generated by an additional unit of capital. Since marginal q is not empirically observable, different substitutes such as
ip t
Tobin’s q (ratio of firm's average value to its capital stock) are introduced in empirical studies. These empirical results show that the explanatory power of Tobin's q in investment equations is
cr
much weaker compared to financial variables. A possible reason for this failure would be the
us
inadequacy of Tobin's q in capturing investment opportunities due to measurement errors.4 Alternative measures of fundamentals are introduced in empirical studies based on non-
an
convex capital adjustment cost models.5 Two examples are profitability shocks and the gap measure between the desired and actual capital stock. In this paper, these two fundamentals are
M
used as proxies for investment opportunities. One big advantage of these fundamentals is that,
d
even though they are constructed using current variables, they present a forward-looking
te
behavior. Since profitability shocks are serially correlated, the current value of shocks gives information about future profitability. The gap measure between the desired and actual capital
Ac ce p
stocks is also informative about future investment behavior, because the magnitude of the gap determines whether a firm invests in the current period or in the future. Another advantage of these fundamentals is that since they give information about the future, the correlation between the fundamentals and current internal funds is low. For example, the correlation between the profitability shocks and the cash flow to capital ratio is only 0.25 as shown in Bayraktar (2002a). 4
The following papers are some of the examples focusing on measurement errors associated with fundamentals:
Kaplan and Zingales (1995 and 1997), Gilchrist and Himmelberg (1995 and 1998), Gomes (2001), Erickson and Whited (2000), Cooper and Ejarque (2001), and Abel and Eberly (2003). 5
See Caballero, Engel and Haltiwanger (1995), Cooper and Haltiwanger (2006), Cooper and Ejarque (2001),
Bayraktar (2002b), Bayraktar (2009), and Bayraktar et al. (2005).
10 Page 10 of 60
4 Data and Construction of Fundamentals The main data source is the COMPUSTAT firm-level database. The data set, covering the period
ip t
from 1981 to 1996, includes U.S. manufacturing firms.6 The total number of firms is 463 and the total number of panel observations is 6450, where the balanced panel would have had 6482
us
cr
observations.7 The following sub-sections introduce main variables used in the study.
4.1 Investment, Financial Variables, and Firm Characteristics
an
The definition of capital includes plant, property, and equipment (PPE). Investment is defined as capital expenditure net of capital sales, including capital retirements. The replacement value of
M
capital is calculated using a perpetual inventory method: Kt = (1-δ) Kt-1 + It, where Kt is the real capital stock, It is real investment, which is calculated by deflating the nominal value by the 4-
d
digit investment price index. δ is the 2-digit depreciation rate from the Bureau of Labor Statistics
te
(BLS) database. It is equal to the average value of the depreciation rates for the period of 1981-
Ac ce p
1996. The investment rate is defined as the ratio of real investment to the replacement value of capital. Since investment is net of sales of capital, there are also negative investment rates, corresponding to nearly 10 percent of observations. The following financial variables capture the effects of internal funds on investment: cash flow to capital ratio (ratio of the book value of cash flow to the beginning of period book value of gross total PPE); sales to capital ratio (ratio of the book value of net sales revenue to the beginning of period book value of PPE); working capital to capital ratio (ratio of the book value 6
Since the retirement data, used in constructing investment series, which were not reported since 1996, the
following years are not included in this study. Some other variables used in constructing the fundamentals are not available for more recent years. 7
Detailed information on sample selection is in Appendix A.
11 Page 11 of 60
of working capital –the difference between current assets and liabilities– to the beginning of period book value of PPE. In the literature different firm characteristics are used in determining firms with possible
ip t
financial restrictions. The following a priori proxies are used in identifying financially constrained firms in this paper: book value of PPE stock; number of employees; total debt to
cr
capital ratio (ratio of the book value of total debt to the beginning of period PPE); bond rating
us
(the issuer’s senior debt rating assigned by Standard & Poor's); dividend payout (ratio of the book value of dividends to the book value of income before extraordinary items); dividend to
an
capital ratio (ratio of the book value of dividends to the beginning of period book value of PPE). In addition to these a priori proxies, an index measure of equity dependence of firms (the KZ
M
index), which is initially introduced by Kaplan and Zingales (1997), is applied in classification of
te
4.2 Fundamentals
d
firms.
Ac ce p
Two different measures of fundamentals are introduced in the paper: profitability shocks and the mandated investment rate. Besides these fundamentals, Tobin's q is also included for the purpose of comparison.
4.2.1 Profitability Shocks
The first set of fundamentals capturing investment opportunities is idiosyncratic profitability shocks. Cooper and Haltiwanger (2006) display that the empirical relationship between the investment rate and profitability shocks is nonlinear and asymmetric such that the response of investment to positive shocks is much stronger than its response to negative shocks which can be explained with the presence of both convex and nonconvex adjustment costs. The profitability shocks are presented in the following firm-level profit function: 12 Page 12 of 60
(4) where
is the profitability shock, consisting of both aggregate and idiosyncratic components,
is the curvature of the profit function, and .
ip t
alternative ways to calculate
is the firm level capital stock. There are two
by regressing the log of real profits on the log of real capital using firm-level panel
us
calculates
The first method
cr
Profitability shocks as residuals from a profit function regression
data. Fixed effects are removed and time dummies are included to identify the effects of
an
aggregate profitability shocks.
ln (.) ln Kit F T rait ,
M
(5)
where rait (firm-level error term) is taken as idiosyncratic profitability shocks and named as
d
residual profitability shocks (rait) from now on.
A second, indirect,
is to consider the first order condition for profit maximization with
Ac ce p
method of calculating
te
Profitability shocks as residuals from first order conditions
respect to employment. Since the employment series are more reliable, this allows us to avoid possible measurement errors in profit data. It is assumed that a firm maximizes the following profit function with respect to labor:
( A, K ) max R( Aˆ , K , L) Lw, L
(6)
where A is the profitability shock that contains both aggregate and idiosyncratic shocks, Aˆ is the shock to the revenue function, K is the capital stock, L is labor, and w represents labor wage. It should be noted that these shocks have been calculated for each firm and period. In order to
13 Page 13 of 60
simplify the equations, time and firm subscripts are removed. Assuming that the product market is imperfectly competitive, the revenue function is defined as
R( Aˆ , K , L) Aˆ py Aˆ y y Aˆ y1 ,
ip t
(7)
where the constant returns to scale Cobb-Douglas production function is assumed to be
cr
y AK K L L , where αL and αK are the production function coefficients for labor and capital,
us
respectively. The demand curve is p y , where ξ is the elasticity of the demand curve.
an
From the first order condition with respect to L, the optimum value of L is:
1
L (1 )
K
K (1 ) 1 L (1 )
. (8)
M
Aˆ (1 ) L L* w
After plugging this optimum labor back into the profit equation, it becomes:
te
d
( A, K ) ( A1 A2 ) K ,
(9)
L (1 )
1
Ac ce p
Aˆ (1 ) L 1 L (1 ) Aˆ (1 ) L 1 L (1 ) (1 L )(1 ) , A2 w . where , A1 Aˆ w w 1 L (1 ) Note that ( A1 A2 ) is the profitability shock, A, which is equal to L (1 ) L (1 ) 1 1 1 L (1 ) 1 L (1 ) 1 L (1 ) 1 L (1 ) ˆ A A w [(1 ) L ] [(1 ) L ] .
(10)
From Equation (8) Aˆ is calculated as
w Aˆ (1 ) L
L* K
L (1 )
. (11)
14 Page 14 of 60
In terms of calibration, is estimated by regressing the log of real net profit data on the log of the replacement value of real capital using firm-level panel data after removing fixed effects. Its value is estimated at 0.61. αL is estimated as 0.73 again using firm-level data.8 ξ is obtained (1 L )(1 ) . These coefficients imply a demand elasticity of -0.15 and a markup of 1 L (1 )
ip t
from
cr
about 18 percent.
us
The aggregate shocks are calculated as the annual mean of A across firms, and the firm-level idiosyncratic component of A, presented by ait, is taken as the deviation from this mean. ait is
an
named, from now on, as profitability shocks from the first order condition. The correlation
M
coefficient between rait and ait is calculated as 0.55 in Bayraktar (2002a).
4.2.2 Mandated Investment Rate
d
Caballero and Engel (1994) try to explain the lumpy nature of investment using a model based
te
on the standard (S,s) literature. They measure imbalances in capital as the gap between the
Ac ce p
desired and the actual capital stocks (mandated investment rate). In their model, the investment rule is that once the measure of imbalance reaches a threshold value, the capital adjustment occurs at once. Firms will often wait until they reach the trigger point due to the presence of nonconvex adjustment costs. This model is empirically studied by Caballero et al. (1995) and Goolsbee and Gross (1997).
The mandated investment rate in a firm at period t, zit, is
8
αL is estimated using cost shares. The cost share of labor is calculated as the ratio of wages times employment to
the sum of the rental price of capital times capital level and wages times employment level. The employment and capital data are from COMPUSTAT database. While the rental price of capital is obtained from BLS database, the wage data are from Gray and Bartelsman's 4-digit dataset.
15 Page 15 of 60
zit ≡ where
- k-1,
(12)
and k-1 represent the natural log of the desired and the actual capital stocks in a firm at
time t, respectively. It should be noted that this measure is calculated for each firm and period;
ip t
time and firm subscripts are removed to simplify the equations. While the positive values of zit correspond to capital shortages, the negative ones indicate excess capital. Desired capital refers
cr
to the stock of capital that a firm would hold if adjustment costs were momentarily removed. It is
=
+ d,
:
an
proportional to the log of the frictionless capital stock,
us
constructed under the following assumptions. First, it is assumed that desired capital is
(13)
hold if it had never faced adjustment costs.
M
where d is a firm-specific constant.9 Frictionless capital is the stock of capital that a firm would
d
The second assumption is that the frictionless stock of capital, , is determined by a
te
neoclassical model. In this model, all frictions are assumed to be absent, including time-to-build
Ac ce p
assumption, or any adjustment costs. Y represents the value of output of an individual firm, assuming imperfect competition, and fixed factors other than capital producing decreasing returns:
(14)
where A, K, and L are profitability shocks, capital stock, and flexible factors, respectively. the capital share of production and
is
represents the flexible factor.
Optimizing over flexible factors yields a profit function: (15) 9
Bertola and Caballero (1994) show that this assumption is compatible with the behavior of a rational firm whose
profit function is isoelastic, and that faces shocks that have independent increments.
16 Page 16 of 60
where
is the price of flexible factors. Given the equation, frictionless capital is defined as (16)
ip t
where c is the cost of capital. After some manipulations and taking the logarithm of the above
(18)
cr
expression, the frictionless capital level is expressed as
where
, and y is the log of real output, which is the sum of the real
is a decreasing function of the curvature of the profit function with respect to
an
of real capital.
us
sale value of goods plus changes in the real value of finished goods inventories, and k is the log
capital, which is approximately equal to 1/(1-α) where α is the cost share of capital, estimated at , where
is the real interest
M
the 4-digit industry level.10 The cost of capital, c, is
rate, which is equal to the average nominal Baa corporate bond rate minus the measure of
d
expected inflation from the Livingston Survey of twelve-month inflation expectations.
is the
te
depreciation rate taken from the 2-digit unofficial BLS data set. In this set, depreciation rates are
Ac ce p
given for three asset groups for each 2-digit industry. The wealth share of assets is used as a weight to calculate the average depreciation rate in each 2-digit industry group. capital expenditure deflator and
is shipments price deflator taken from Gray and Bartelsman
data set (at the 4-digit industry level). The tax parameter,
where and 10
is the new
is the corporate income tax,
, is taken from the BLS database: (19)
is the present value of $1 of tax depreciation allowances,
is the effective rate of investment tax credit. This equation is given for 93 assets in each
= (rental price of capital*capital stock)/total cost of production. It is calculated at the firm-level. The data source
for rental price is BLS database and for total cost of production is COMPUSTAT. Its average value is 1.37.
17 Page 17 of 60
two-digit industry level. In order to find the two-digit weighted average values, the wealth share of assets in each industry is calculated. The average value of c is 0.16.
(20)
is the long-run elasticity of capital with respect to its cost in a firm at time t.
cr
where
ip t
With slight modifications, the empirical equation for the firm-level gap measure is:
The third and last assumption in calculating mandated investment is the estimation of
from
us
a cointegrating regression of the natural log of the capital-to-output ratio on the cost of capital at the 2-digit industry level using firm-level panel data.11 This coefficient can be interpreted as the
an
long-run elasticity of capital with respect to its cost. Bayraktar (2002a) calculates the average
M
value of this measure as -0.68. This value is close to -1, which is the long-run elasticity in neoclassical investment models.
d
After calculating frictionless capital, the firm-specific constant, d in Equation (13), is and
for the five points with investment
te
estimated by taking the average gap between
Ac ce p
closest to median investment, which can be thought of as maintenance investment. 4.2.3 Tobin’s q
Tobin's q, the ratio of the market value of firms to the replacement value of capital, is the most commonly used fundamental determinant of investment in the literature. In this study, Tobin's q is calculated following Barnett and Sakellaris (1998).12 The numerator is the sum of the market value of common stock, the liquidating value of preferred stock, the market value of long-term
11
The twice-lagged first difference of the log of cost of capital is included as explanatory variable in order to reduce
the small sample bias. The data source is COMPUSTAT. 12
Details on the calculation of this fundamental are given in Appendix B.
18 Page 18 of 60
debt and the book value of short-term debt. The denominator is the sum of the replacement value of fixed capital and inventories.
ip t
5 Empirical Results
cr
The main aim of this section is to investigate empirically systematic differences between financially constrained and unconstrained firms in their investment sensitivity to fundamentals.
us
In order to accomplish this purpose, different investment regression specifications are run first
an
for constrained versus unconstrained firms as defined below. Sensitivities of investment to fundamentals are measured by the estimated coefficients of these investment regressions. Then,
M
as reported in section 5.5, the systematic differences in investment responses to the different fundamentals are calculated across firm groups by using the estimated coefficients and shocks to
d
the fundamentals as observed in actual data.
te
In this section the following reduced form investment equation is estimated by a least
Ac ce p
squares regression technique for panel data:
(21)
where i represents firms and t represents years.13
is the investment rate at firm i in period t.
is a fundamental measure capturing investment opportunities: two types of profitability shocks (rait and ait), the mandated investment rate (zit), or Tobin's q.14 Since Bayraktar (2002a) shows that since the relationship between the investment rate and these fundamentals is 13
For the purpose of comparison, empirical specifications only with the linear term of the fundamentals are also run
and the results are reported in the regression tables. 14
It should be noted that despite the fact that Tobin’s q possibly involves measurement errors as explained in the
previous sections, it is still included in the regression specifications for the purpose of comparison.
19 Page 19 of 60
nonlinear, the squared term of the fundamentals is also included.15 FV controls for financial variables: the cash flow to capital ratio, the ratio of sales to capital, or working capital to capital.16 T represents a set of time dummies, and F a set of firm dummies to remove fixed
ip t
effects. u is the error term.
A regression specification with FV as the only independent variable is also run to understand
cr
the impact of FV on investment when fundamentals are excluded. This third specification can
us
give better information on how estimated coefficients of FV change when fundamentals are added to the empirical equation. Given that both x and FV are expected to be strong determinants
an
of investment, their joint effect can also have a significant effect on investment. In a fourth specification, an interaction term between FV and x is added in Equation (21) to analyze whether
M
or not the effect of x on investment depends on FV.
d
Each empirical specification is run separately for financially constrained versus
te
unconstrained firms. In order to identify financially constrained firms as defined in Section 4.1, the firms in the data set are split into sub panels using six alternative a priori criteria and the KZ
15
Ac ce p
index of equity dependence of firms as explained in Section 4.1.17
Related to the nonlinear relationship between fundamentals and investment, also see Barnett and Sakellaris
(1998), Barnett and Sakellaris (1999), and Abel and Eberly (2002). 16
All these ratios are calculated using the book values. The results do not change when the real values are used
instead. In addition to the financial variables mentioned above, the ratio of cash and equivalence to capital and the ratio of tax payments to income are also included. The results are not reported here because the ones with the cash and equivalence to capital ratio were similar to the results with the ratio of cash flow to capital, and the statistical significance of tax payments was negligible. 17
Other criteria such as total assets, real capital stock, flow and stock of debt burdens, and interest coverage ratio are
also introduced. But since they produce similar results, they are not reported in the paper.
20 Page 20 of 60
In subsection 5.5 the joint effects of the linear and the squared terms of fundamentals on investment are systematically investigated. The hypotheses for the results presented in that subsection are:
ip t
H0: investment/fundamental sensitivities are higher for financially constrained firms than sensitivities of unconstrained firms;
cr
H1: investment/fundamental sensitivities are not higher for financially constrained firms than
us
sensitivities of unconstrained firms.
The regression specifications used in the analysis also allow us to test the sensitivity of
an
investment to financial variables as well as the sensitivity of investment to the linear and squared terms of fundamentals separately across different firm classifications. In these tests, the statistical
M
significance of the differences in the estimated coefficients of financial variables across firm
d
groups and the differences in the estimated coefficients of the linear and the squared terms of
te
fundamentals across firm groups are calculated. The test results are reported in Tables 2 to 8. As
are:
Ac ce p
the coefficients are defined in equation (21), the null and alternative hypotheses of these tests
H0: b1 for constrained firms – b1 for unconstrained firms = 0; H1: b1 for constrained firms – b1 for unconstrained firms ≠ 0; and
H0: b2 for constrained firms – b2 for unconstrained firms = 0; H1: b2 for constrained firms – b2 for unconstrained firms ≠ 0; and H0: b3 for constrained firms – b3 for unconstrained firms = 0; H1: b3 for constrained firms – b3 for unconstrained firms ≠ 0.
21 Page 21 of 60
In the following subsections, the results are presented where firms are grouped according to: 1) the size of their capital stocks and the number of employees; 2) the level of dividend ratios; 3) the level of their debt stock and the availability of bond rating; and 4) the KZ index. After these
ip t
analyses, in the last subsection, the joint effects of linear and nonlinear terms of the fundamentals on investment are reported to summarize the systematic differences in investment/fundamental
us
cr
sensitivities between financially constrained and unconstrained firms.
5.1 Size of Firms
an
The first two criteria to classify financially constrained and unconstrained firms are the size of firms in terms of their capital stocks and the number of employees. Small firms (Class 1) are
M
defined as those with the average capital stock (or number of employees) in the bottom half of the empirical distribution. Since firms with a low capital stock typically have fewer employees,
d
the results with these two different classifications are very similar. One would expect smaller
te
firms to be financially constrained, since costs of getting external funds are presumably higher
Ac ce p
for them, and public information about their investment projects is generally more limited. These facts restrict their ability to find external funds. Given that small firms are expected to be financially constrained, their investment should respond more to changes in fundamentals or investment opportunities, because improving fundamentals indicate an improving financial position through declining agency costs. The average values of the variables for these two classifications are reported in Columns (3) to (6) in Table 1. The average investment rates in groups are similar. Smaller firms grow faster, retain a higher fraction of their income, and pay fewer dividends as a proportion of their income. When the average values of the fundamentals are compared, they are very close for the profitability shocks and the mandated investment rate. Table 1 also reports that the correlation 22 Page 22 of 60
between profitability shocks and investment is higher for small firms; they are very similar when zit is the fundamental measure and substantially low for small firms when Tobin’s q is the fundamental measure.
ip t
The estimation results based on two classifications are presented in Table 2 (size of capital stock) and Table 3 (number of employees). In almost every case, while the coefficient of the
cr
linear term of the fundamentals is larger for small firms, the coefficient of the squared term is
us
higher for large firms. The difference between the estimated coefficients of the two groups of firms is statistically significant as reported in Tables 2 and 3. The results are robust whether the
an
sample is split by the size of capital stock or by the number of employees. One possible explanation of the findings would be that smaller firms may have lower non-convex capital
M
adjustment costs, such as fixed costs, due to their smaller size. As a result, these firms can
d
linearly follow investment opportunities. Another possible explanation is that the higher
te
sensitivity of smaller firms’ investment to the shocks to the fundamentals as expected by the contracting models of financial market imperfections.18 Firms with financial constraints are more
Ac ce p
sensitive to the fundamentals (profitability shocks or mandated investment), since these fundamentals signal investment opportunities, as well as changes in the financial position. Another interesting observation is that when the relative magnitudes of the estimated coefficients are compared across class 1 and class 2 firms in regression specifications where fundamentals and financial variables are introduced together, smaller firms' investment tends to be less sensitive to changes in financial variables, even though the opposite is expected
18
The joint effects are reported in Section 5.5.
23 Page 23 of 60
according to the mainstream financial market imperfection literature.19 The test results presented in Tables 2 and 3 confirm these observations. The exception is Tobin's q, for which the coefficients of financial variables are higher for smaller firms, as presented in many empirical
ip t
studies.
Another result is that in the investment regression specifications, where only financial
cr
variables are included as independent variables, it can be seen that the magnitude of the
us
estimated coefficient of the financial variable is larger for smaller firms. But the size of the coefficient drops sharply for this group of firms, and even becomes insignificant in some
19
M
an
regression specifications, when fundamentals are introduced in the empirical model.
The high sensitivity of investment to the cash flow ratio for financially unconstrained firms is also observed in the
d
study by Kaplan and Zingales (1997). The least constrained and the most financially successful firms in their sample
te
seem to depend primarily on their cash flow to finance their investment despite the availability of additional low
Ac ce p
cost funds. In their study, they use Tobin's q to control investment opportunities, and classify firms as financially constrained by undertaking an in-depth analysis of firms. Erikson and Whited (2000) also show that cash flow does not matter, even for financially constrained firms once measurement errors are corrected. Baker, Stein, and Wurgler (2003) also report the higher sensitivity of investment to fundamentals and financial variables for highly equitydependent firms, which are identified following the methodology suggested by Kaplan and Zingales (1997). The main differences between their and this paper is that, in their analysis they do not include any nonlinear terms of fundamentals which are important determinants of investment behaviors of firms. While they use different investment measures but only one fundamental variable (Q) and one financial variable (cash flow-to-assets ratio), four different measures of fundamentals and three different financial variables are tested in this paper. Moreover, they use only the KZ index to identify financially constrained firms, while seven different classifications, including the KZ index, are used in this paper. Tsoukalas (2011) shows that time-to-build for capital projects create an investment-cash-flow sensitivity that may not be indicative of capital market frictions.
24 Page 24 of 60
There can be different reasons for why more financially constrained firms exhibit lower investment-internal funds sensitivity compared to unconstrained ones. As specified by Kaplan and Zingales (1997 and 2000), one possible reason could be excessive conservatism by
ip t
managers, which may be caused by the way firms are organized or by non-optimizating behavior of managers. Alternatively, as predicted by the contracting models, improvements in
cr
fundamentals would be more important for the investment decision of financially constrained
us
firms when compared to any improvements in internal funds alone. This is because their financial constraints start relaxing more with good fundamental shocks due to a declining agency
an
cost of borrowing.
Given the statistical and economic significance of fundamentals and financial variables in
M
determining fixed investment, a regression with an interactive term between financial variables
d
and fundamentals is included to understand their joint impact on investment. In Tables 2 and 3 it
te
can be seen that, in general, the coefficients of the interaction term are positive and statistically significant for smaller firms, while it is negative and significant or positive but insignificant for
Ac ce p
larger firms. The trend is robust across different financial variables, fundamentals, and firm classifications. The positive and significant coefficient of the interactive term for smaller firms means that at higher values of fundamentals, the impact of financial variables on investment gets larger; or at higher values of financial variables, the impact of fundamentals on investment gets larger. As a result, for smaller firms the effect of financial variables on investment positively depends on fundamentals, and at the same time, it suggests that the effect of fundamentals on investment depends on financial variables. Given that smaller firms tend to be more financially constrained, the positive and significant interaction term reinforces the argument that the joint impact of the two determinants of investment (fundamental and financial) are essential in the
25 Page 25 of 60
investment process of financially constrained firms. On the other hand, for financially less constrained firms the coefficient of the interaction term is negative, or positive but not significant. For negative and significant coefficients, it can be concluded that at higher values of
ip t
fundamentals the impact of financial variables on investment gets smaller for less constrained firms. In this case, the effect of fundamentals on investment depends negatively on financial
cr
variables. The two determinants of investment do not reinforce each other in their fixed
us
investment activities. Given that these firms are expected to be less financially constrained, it is
an
an expected result.
5.2 Size of Dividends
M
The third and the fourth criteria used in identifying financially constrained firms are the dividend payout ratio and the dividend to capital ratio, respectively. Firms are classified as financially
d
constrained (Class 1) if their ratios are in the bottom half of the empirical distribution. Class 2
te
represents groups with higher dividend payments. One would expect Class 2 to be financially
Ac ce p
unconstrained since they pay a large fraction of their income as dividends, which suggests that the opportunity cost of their internal funds is low (Fazzari et al., 1988). The average values of the variables for all groups are reported in Columns (7)-(10) in Table 1. The capital stock and the number of employees are lower for Class 1 firms; but they grow faster and retain a larger fraction of their income. While the average investment rate is 0.15 for Class 1 firms, it is 0.11 in Class 2. The simple correlations between investment and fundamentals indicate that the sensitivity of investment for high dividend paying firms to the fundamentals is stronger. One explanation for this unexpected result can be that high-dividend paying firms would actually be more dependent on fundamentals to be able to more easily finance their
26 Page 26 of 60
investment, given that they distribute a higher proportion of their internal funds as dividends. In this case, it looks like higher-dividend payers are more financially constrained. However, this observation which is based on a simple correlation analysis changes with
ip t
regression analysis. The estimation results are presented in Tables 4 and 5. In reviewing Table 4, in general the sensitivity of investment to the linear term of the fundamentals is higher for the
cr
firms with low dividend payouts, while the sensitivity to the squared term is higher for the firms
us
with high dividend payouts. As presented in the tables, the difference between the estimated coefficients of the two groups of firms is statistically significant in most cases. There are some
an
exceptions to this trend. When rait is the fundamental measure of investment and the sales-tocapital ratio or working capital-to-capital ratio are the financial variables, the linear term of
M
fundamentals is smaller for the financially constrained group. However, in these cases the
d
difference between the coefficients of the two groups of firms is not statistically significant.
te
As was the case with smaller firms, those with lower dividend payouts seem to follow fundamentals linearly. Again the results support the expectations of contracting models. Given
Ac ce p
that the magnitude of the nonlinear term in some results is much higher for the financially unconstrained firms, the joint effect of linear and nonlinear terms of the fundamentals on investment is expected in some cases to be close for the financially constrained and unconstrained groups. This is further discussed in Section 5.5. In Table 5, we observe the trend that financially constrained firms’ investment responds more to the linear term of fundamentals, but less to the squared term, only when zit is the fundamental measure. When ait is the fundamental determinant of investment, the size of both linear and squared terms is larger for financially constrained firms. As a result, ait is predicted to have a stronger impact on investment when firms are classified based on the level of dividends-to-
27 Page 27 of 60
capital ratio. For rait and q, the results are mixed. It is worth noting that when rait is the fundamental determinant, the difference between the estimated coefficients of Class 1 firms and Class 2 firms is not statistically significant in most cases.
ip t
Even though Class 1 firms are expected to be financially constrained, the sensitivity of the investment rate to the cash flow ratio across firm classifications is lower compared to its
cr
sensitivity to the fundamentals, except in the case of Tobin’s q, which may have a possible
us
mismeasurement problem. As was the case in Tables 2 and 3, for expected-to-be financially constrained firms, the coefficients of the fundamentals are robust to the inclusion of financial
an
variables in the regression, while the coefficients of the financial variables drop significantly when they are included together with fundamentals in the same equation. The results also show
M
that the interaction terms between financial variables and fundamentals again tend to have a
d
positive and significant effect for the constrained group, while it usually has a negative and
te
significant or positive but insignificant effect on the unconstrained group.
Ac ce p
5.3 Size of debt to capital ratio and bond rating The last two criteria used in identifying financially constrained firms are the availability of a bond rating and the size of the debt to capital ratio. Since the debt burden of firms with a high debt ratio or without any bond rating (Class 1) is heavier, they are expected to be more financially constrained. Therefore, their investment should be more sensitive to changes in internal funds as expected by the mainstream financial market imperfections literature, or alternatively it is expected to be more sensitive to the fundamentals as expected by the contracting models of financial frictions. Firms with a bond rating are expected to be less financially constrained because public debt issuance is a good indicator of firms' financial position, given that it provides a low-cost access to capital markets (Whited, 1992; Calomiris et 28 Page 28 of 60
al., 1994; Agca and Mozumdar, 2008). Firms with a higher debt burden are expected to have more financial problems because they are expected to pay higher premiums to obtain external funds (Bernanke et al., 1990; Whited, 1992; Hu and Schiantarelli, 1998; Hennessy et al., 2007).
ip t
Columns (11)-(14) in Table 1 show the average values of the variables for the firms classified based on these two criteria. The share of firms with a bond rating is 0.30. The capital
cr
stock and the number of employees of firms without a bond rating are lower. Firms with a low
us
debt to capital ratio have a larger amount of capital on average, but they have fewer employees. The average investment rate of firms with a high debt to capital ratio is slightly higher. The
an
simple correlation between investment and the fundamentals tends to be stronger for the financially constrained firms except in the case of Tobin’s q.
M
Tables 6 and 7 exhibit the estimation results. In Table 6 the investment rate of firms without
d
a bond rating or with a high debt to capital ratio tends to respond more to changes in the linear
te
term of the fundamentals, but less to the squared term, as was the case with other classifications. This observation is true even in the specifications with Tobin’s q. For constrained firms (firms
Ac ce p
without a bond rating), the dropping economic significance of financial variables in determining investment in the presence of fundamentals can supports the argument that fundamentals can capture impacts of financial variables on investment. Table 7 shows that the sensitivity of the investment rate to both financial variables and fundamentals is stronger for the firms with a high debt to capital ratio (Class 1) when compared to the sensitivity of Class 2 firms. The exception is that when zit is the fundamental determinant of investment, both linear and squared terms of fundamentals are statistically significantly higher for constrained firms. It is important to note that the size of the financial variable tends to be higher for Class 1 firms. This is an expected result since these firms have a heavier debt burden.
29 Page 29 of 60
Consequently, any changes in financial variables or fundamentals should have a larger effect on the investment rate. In Table 7, when the cash-flow ratio is the financial variable in the regression, the explanatory power of this financial variable decreases for the firms with a high
ip t
debt ratio relative to the explanatory power of the same financial variable for the firms with lower debt burden. However, the significance of the fundamentals continues in the same
cr
equations for Class 1 firms.
us
As was the case in other classifications, it is observed that the estimated coefficients of the interaction terms between financial variables and fundamentals are positive and significant for
an
constrained firms, while they are mostly negative or insignificant for firms with bond rating or
M
lower debt burden.
d
5.4 Firm Classification Based on the KZ index
te
The KZ index is initially introduced by Kaplan and Zingales (1997) to identify financially constrained firms. They name potentially financially constrained firms as equity-dependent
Ac ce p
firms. Using this classification, Kaplan and Zingales (1997) show that equity-dependent firms are more sensitive to fundamentals. While constructing this index, Kaplan and Zingales undertook a detailed study of the financial constraints of 49 low-dividend manufacturing firms. Based on different subjective and objective criteria, they sorted these firms from least to most constrained. After creating this ranking, they ran a regression to link this new index to the series from COMPUSTAT so that the index could be used in other studies. Lamont, Polk, and SaaRequejo (2001) and Baker, Stein, and Wurgler (2003) are two prominent examples of empirical studies which construct the KZ index using the COMPUSTAT database. In this paper, following
30 Page 30 of 60
the methodology of Baker, Stein, and Wurgler (2003), the KZ index is created using the
where KZ is the index;
is the ratio of cash flow (items 14 and 18 in COMPUSTAT) to
is the ratio of dividends (items 21 and 19) to assets;
is the ratio of cash
cr
assets;
ip t
following equation in which the estimated coefficients come from Kaplan and Zingales (1997):
is leverage (items 9+34 divided by items 9+34+216). The
us
balances (item 1) to assets; and
an
higher the index value, the higher equity dependence is. When the firms are ranked based on their average index value from low to high and split into two groups, firms in the high value
M
group are expected to be equity dependent or financially constrained (Class 1), and firms with lower index values are considered to be financially unconstrained (Class 2).
d
Table 1 (columns (15) and (16)) reports the descriptive statistics for these two groups of
te
firms. Equity-dependent firms tend to be smaller both in terms of their capital stock and the
Ac ce p
number of employees. They tend to invest slightly less. While their dividend payout is lower, their debt ratios are higher. The correlation between the investment rate and the fundamentals are higher for equity-dependent firms, except when q is the fundamental determinant of investment. Table 8 reports the estimated coefficients for different specifications and firm classifications. The results support the previous findings. For equity-dependent firms, which are expected to be financially constrained, both the linear and squared terms of the fundamentals are higher when rait and q are the fundamental determinants. However, the squared term is lower for them when the fundamentals are measured by ait and zit. In section 5.5, the joint effects of the linear and squared terms of the fundamentals are systematically investigated.
31 Page 31 of 60
In general, the estimated coefficients of the financial variables are statistically significantly higher for equity-dependent firms than the ones for equity-independent firms. The sign, significance, and the magnitude of the interaction terms between the financial variables and
cr
ip t
fundamentals indicate a strong, positive joint effect on investment for Class 1 firms.
5.5 Joint Effects of Linear and Nonlinear Terms of Fundamentals on Investment
us
For the most part, regression results indicate that the coefficient of the linear term of the fundamentals is higher for the financially constrained firms, while the coefficient of the squared
an
term is higher for the financially unconstrained firms. These findings are robust across
M
alternative classifications used in distinguishing the financially constrained firms and across alternative methods to calculate the fundamentals. Before any conclusion is made, it is important
d
to assess for which group of firms the joint effects of the linear and nonlinear terms are stronger
te
in the sample used for estimation. In this assessment the estimated coefficients of the linear and
used.
Ac ce p
squared terms of the fundamentals and the standard deviations of the fundamental shocks are
For these analyses, different shock values of each fundamental are calculated and reported in the first row of Table 9. The shock values are within ±1 or ±2 standard deviations of the shocks. A set of estimated coefficients are needed to calculate the joint effects. Given that the results are mostly robust to which financial variable is included in the specifications, and given that the cash-flow to capital ratio (CF_K) is one of the most commonly used financial variables in the literature, the following regression equation is picked to construct the joint effects of the linear and nonlinear terms of the fundamental shocks: (22) 32 Page 32 of 60
where x stands for the fundamentals ait, rait, or zit. The estimated coefficients of this specification are reported in Columns (6) and (10) of Tables 2 to 8. While the results in Column (6) are for the financially constrained firms, the ones in Column (10) are for the unconstrained where
ip t
firms. The values in Table 9 are calculated as:
coefficients of the linear and nonlinear terms of the fundamentals.
cr
the shock values are reported in the first row of the table and b1 and b2 are the estimated
us
The joint effects of the linear and the squared terms of the fundamentals are substantially stronger for the financially constrained firms within the ±2 standard deviation limits of the
an
shocks. The asymmetry between the effects of the negative versus positive shocks is substantial. The firms in the financially constrained groups respond more strongly to the positive shocks
M
when compared to the responses by the unconstrained firms. The financially constrained firms
d
react even more strongly to the negative shocks by cutting their investment significantly. There
te
are only a few exceptions where the response of investment to the positive shocks is slightly higher for the unconstrained firms. All of these exceptional cases belong to the classifications
Ac ce p
based on dividend payouts or the ratio of dividends to capital. Overall, when a realistic size of a shock is assumed, the estimated linear and nonlinear coefficients of the fundamentals imply that financially constrained firms invest more when a positive shock realizes, and cut their investment more significantly when a negative shock occurs.
6 Conclusion The aim of this paper is to investigate whether or not the empirical and economic significance of fundamentals, measured by profitability shocks and mandated investment rate, are higher in determining fixed investment behavior of financially constrained firms, as is predicted by the 33 Page 33 of 60
contracting models of financial frictions. The main point is that fundamentals are indicators of improving investment opportunities, as well as easing financial constraints through dropping agency costs of borrowing. Therefore, investment by financially constrained firms is expected to
ip t
respond more strongly to changes in fundamental determinants of investment. The findings based on regression analyses support this expectation. One common finding is that the
cr
coefficient of the linear term of the fundamentals in investment regressions tends to be higher for
us
financially constrained firms, while the coefficient for the nonlinear term of the fundamentals tends to be higher for financially unconstrained firms. But the joint effects of the linear and
an
nonlinear terms of the fundamentals within ±2 standard deviation of the shocks are stronger for financially constrained firms whether shocks are positive or negative. These results are quite
M
robust across different methods to construct the fundamentals and different classification
d
methods to identify the financially constrained firms. This would be a useful finding for
its critics.
te
researchers since it cannot be easily explained by the framework of Fazzari et al. (1988), nor by
Ac ce p
The results show that financial variables are still statistically significant determinants of firmlevel fixed investment. But the interesting outcome is that in most cases the explanatory power of the financial variables in the investment process tends to be lower for expected-to be financially constrained firms when compared to the explanatory power of the same financial variables for the unconstrained group of firms. This finding further supports the argument that constrained firms are more dependent on the changes in fundamentals to ease their financial burdens, as expected by contracting models. For the purpose of comparison, the same analyses are repeated using Tobin's q as a measure of fundamental. In regression specifications, in which Tobin’s q is the fundamental, financial
34 Page 34 of 60
variables are relatively more significant in explaining investment compared to the impact of Tobin’s q (fundamental measure) on investment. The result is robust for both constrained and unconstrained firms. These outcomes indicate that the statistical and economic significance of
the inadequacy of Tobin's q in capturing investment opportunities.
ip t
financial variables, but not fundamentals, in explaining investment indeed would be caused by
cr
In terms of future studies related to this topic, the robustness of the results across different
us
industries should be investigated. One possibility is that the analyses may focus only on durable-
Ac ce p
te
d
M
an
goods industries, which are more homogenous, compared to nondurables industries.
35 Page 35 of 60
References Abel, A.B., Eberly, J.C., 2002. Investment and q with Fixed Costs: an Empirical Analysis. Unpublished results, University of Pennsylvania (January).
ip t
Abel, A.B., Eberly, J.C., 2003. Q Theory Without Adjustment Costs and Cash Flow Effects
cr
Without Financing Constraints. Unpublished results, University of Pennsylvania (October).
us
Agca, S., Mozumdar, A., 2008. The impact of capital market imperfections on investment–cash
an
flow sensitivity. Journal of Banking & Finance 32, 207–216.
Albuguerque, R., Hopenhayn, H., 2004. Optimal Lending Contracts and Firm Dynamics. Review
M
of Economic Studies 71(2), 285-315.
Baker, M., Stein, J. C., Wurgler, J., 2003. When Does the Market Matter? Stock Prices and the
te
969-1005.
d
Investment of Equity-Dependent Firms. The Quarterly Journal of Economics 118(3),
Ac ce p
Barnett, S., Sakellaris, P., 1998. Nonlinear Response of Firm Investment to Q: Testing a Model of Convex and Non-convex Adjustment Costs. Journal of Monetary Economics 42, 261288.
Barnett, S., Sakellaris, P., 1999. A New Look at Firm Market Value, Investment and Adjustment Cost. The Review of Economics and Statistics 81, 250-60. Bayraktar, N., 2009. Investment, Alternative Measures of Fundamentals, and Revenue Indicators. International Journal of Revenue Management 3(2), 148-178. Bayraktar, N., 2002a. Analyses of Alternative Fundamentals of Fixed Capital Investment. Unpublished results, University of Maryland.
36 Page 36 of 60
Bayraktar, N., 2002b. Effects of Financial Market Imperfections and Non-convex Adjustment Costs in the Capital Adjustment process. Unpublished results, University of Maryland.
Investment. ECB Working Paper Series No. 566 / December 2005.
ip t
Bayraktar, N., Sakellaris, P., Vermeulen, P., 2005. Real versus Financial Frictions to Capital
cr
Bernanke, B.S., Campbell, J.Y., 1988. Is there a Corporate Debt Crisis?. Brookings Papers on Economic Activity 1, 83-139.
us
Bernanke, B.S., Campbell, J.Y., Whited, T., 1990. US Corporate Leverage: Developments in
an
1987 and 1988. Brookings Papers on Economic Activity 1, 255-286. Bernanke, B.S.; Gertler, M., 1990. Financial Fragility and Economic Performance. Quarterly
M
Journal of Economics (February), 87-114.
Bernanke, B.S., Gertler, M., Gilchrist, S., 2000. The Financial Accelerator in a Quantitative
d
Business Cycle Framework. In Taylor, J.B., Woodford, M. (Eds.), Handbook of
te
Macroeconomics, Vol.3, North Holland.
Ac ce p
Bertola, G., Caballero, R.J., 1994. Irreversibility and Aggregate Investment. Review of Economic Studies 61, 223-246. Brainard, W.C., Shoven, J.B., Weiss, L., 1980. The Financial Valuation of the Return to Capital. Brookings Papers on Economic Activity 2, 453-511. Caballero, R.J., Engel, E.M.R.A., 1994. Explaining Investment Dynamics in the U.S. Manufacturing : A Generalized (S,s) Approach. NBER Working Paper No. 4887. Caballero, R.J., Engel, E.M.R.A., Haltiwanger, J., 1995. Plant-Level Adjustment and Aggregate Investment Dynamics. Brookings Papers on Economic Activity 2, 1-39.
37 Page 37 of 60
Calomiris, C.W., Himmelberg, C.P., Wachtel, P., 1994. Commercial Paper, Corporate Finance, and The Business Cycle: A Microeconomic Perspective. NBER Working Paper, No. 4848 (September).
ip t
Carlstrom, C., Fuerst, T., 1997. Agency Costs, Net Worth, and Business Fluctuations: A
cr
Computable General Equilibrium Analysis. American Economic Review 87, 893-910. Chatterjee, S., Corbae, D., Nakajima, M., Rios-Rull, J-V., 2007. A Quantitative Theory of
us
Unsecured Consumer Credit with Risk of Default. Econometrica 75(6), 1525-89.
an
Clementi, G.L., Hopenhayn, H., 2006. A Theory of Financing Constraint and Firm Dynamics. Quarterly Journal of Economics 121(1), 229-265.
M
Cooley, T., Marimon, R., Quadrini, V., 2004. Aggregate Consequences of Limited Contract Enforceability. Journal of Policy Economics 112, 817-847.
te
Review 91(5), 1286-1310.
d
Cooley, T., Quadrini, V., 2001. Financial Markets and Firm Dynamics. The American Economic
Ac ce p
Cooper, R.W., Ejarque, J., 2001. Exhuming Q: market power vs. capital market imperfections. NBER Working Paper Series; No. 8182 (March). Cooper, R.W., Haltiwanger, J.C., 2006. On the Nature of Capital Adjustment Costs. Review of Economic Studies 73(3), 611-33. Del Boca, A., Galeotti, M., Rota, P., 2008. Non-convexities in the adjustment of different capital inputs: A firm-level investigation. European Economic Review 52, 315–337. Erickson, T., Whited, T.M., 2000. Measurement Error and the Relationship between Investment and q. Journal of Political Economy 108(5), 1027-57.
38 Page 38 of 60
Fazzari, S.M., Hubbard, G.R., Petersen, B.C., 1988. Financing Constraints and Corporate Investment. Brookings Paper for Economic Activity 1, 141-195. Gilchrist, S., Himmelberg, C.P., 1995. Evidence on the Role of Cash Flow for Investment.
ip t
Journal of Monetary Economics 36, 541-572.
cr
Gilchrist, S., Himmelberg, C.P., 1998. Investment, Fundamentals and Finance. NBER Working Paper # 6652 (July).
us
Gomes, J.F., 2001. Financing Investment. American Economic Review 91(5), 1263-85.
an
Goolsbee, A., Gross, D.B., 1997. Estimating Adjustment Costs with Data on Heterogeneous Capital Goods. NBER WP No. 6342 (December).
M
Hennessy, C.A., Levy, A., Whited, T.M., 2007. Testing Q theory with financing frictions. Journal of Financial Economics 83, 691–717.
d
Hennessy, C.A., Whited, T.M., 2007. How costly is external financing? Evidence from a
te
structural estimation. The Journal of Finance 62(4), 1705–1745.
Ac ce p
Hu, X., Schiantarelli, F., 1998. Investment and Capital Market Imperfections: A Switching Regression Approach Using U.S. Firm Panel Data. The Review of Economics and Statistics 80(3), 466-479.
Kaplan, S.N., Zingales, L., 1995. Do Financing Constraints Explain Why Investment is Correlated with Cash Flow?. NBER Working Paper. No. 5267 (September). Kaplan, S.N., Zingales, L., 1997. Do Investment Cash Flow Sensitive Provide Useful Measure of Financing Constraints?. Quarterly Journal of Economics 112(1), 169-215. Kaplan, S.N., Zingales, L., 2000. Investment-Cash Flow Sensitivities are not Valid Measures of Financing Constraints. NBER Working Paper. No. 7659 (April).
39 Page 39 of 60
Lamont, O., Polk, C., Saa-Requejo, J., 2001. Financial Constraints and Stock Returns. Review of Financial Studies XIV, 529-554. Marcet, A., Marimon, R., 1992. Communication, Commitment and Growth. Journal of Economic
ip t
Theory 58(2), 219-249.
cr
Salinger, M., Summers, L.H., 1983. Tax Reform and Corporate Investment: A Microeconomic Simulation Study. In: Martin, F. (Eds.) Behavioral Simulation Methods in Tax Policy
us
Analysis.
Journal of Economic Theory 21, 265-293.
an
Townsend, R., 1978. Optimal Contracts and Competitive Markets with Costly State Verification.
M
Tsoukalas, J.D., 2011. Time to build capital: Revisiting investment-cash-flow sensitivities. Journal of Economic Dynamics & Control 35, 1000–1016.
d
Whited, T.M., 1992. Debt, Liquidity Constraints, and Corporate Investment: Evidence from
Ac ce p
te
Panel Data. Journal of Finance 47(4), 1425-1460.
40 Page 40 of 60
APPENDIX A Sample Selection
Ac ce p
te
d
M
an
us
cr
ip t
The major data source is Standard & Poor's COMPUSTAT Database, which is a database of financial, statistical, and market information covering publicly traded companies. The data set covers only U.S. manufacturing firms with a SIC code between 2000-3999 for the years from 1981 to 1996. The calendar year, which is a period of one year beginning with January 1 and ending with December 31, is taken. In the initial data set, the number of U.S. manufacturing firms was 3798 active firms and 3503 inactive firms, for which data are no longer collected because of such reasons as merger, acquisition, and bankruptcy. The elimination of firms is conducted following several steps: 1. I filter the data depending on the availability of the plant, property, and equipment (PPE) series. I delete firms with missing data points more than two periods for the capital expenditure for PPE and for total gross PPE between 1981 and 1996. All foreign firms are also eliminated. The total number of firms after this elimination is 1200. 2. Then I delete the firms, which have involved in significant acquisitions or merger activity. There are different ways to identify these firms. One way is the elimination of firms, which have expenditure on acquisitions accounted for more than 15 percent of assets. This way has not been used since the data for acquisitions are missing for many firms. The other commonly used way is the exclusion of firms whenever |Gt – Gt-1 – It + Rt| > 0.15Gt-1, where Gt is the book value of the capital stock, It is nominal capital expenditure and Rt is the retirement of capital. This method is taken from Gilchrist and Himmelberg (1995). For some firms, the retirement series are missing. In this case, the retirement value is set to be zero unless the left hand side of the formula is negative. In this case, 10 percent of the previous period's capital is taken as retirement value. The idea behind this formula comes from the accounting identity: the book value of physical capital is equal to the capital level of the previous period plus nominal investment at the current period minus retirement plus the residual balance, which includes changes in capital caused by mergers or acquisitions. So if a firm's residual balance is greater than 0.15 of its previous period capital, then it is assumed that this firm gets into a merger or acquisition activity, and then it is deleted. After applying this method, 500 firms are left. 3. Two firms are eliminated since their data are discontinuous due to FASB 94 (Financial Accounting Standards Board), which requires wholly owned subsidiaries to be included on the firm's balance sheet. The total number of firms drops to 498 after deleting these two firms. 4. The investment series is constructed through subtracting the sale of PPE and retirement of PPE series from the capital expenditure on PPE series. Due to the presence of additional missing data points for the sale of capital and retirement series, the investment rate series could not be constructed for some firms. This fact decreases the number of firms to 471. 5. The replacement value of the capital stock is constructed using a perpetual inventory method as specified in Section 4.1. 8 of firms are deleted since this method produces negative capital stock for them. After all, the number of firms is 463.
41 Page 41 of 60
APPENDIX B Details on the Calculation of Tobin’s q
ip t
The numerator of average Tobin's q is the sum of the market value of common stock, the liquidating value of preferred stock, the market value of long-term debt, and the book value of short-term debt. The denominator is the sum of replacement value of fixed capital and inventories. These variables are calculated as follows:
if INVt* INVt*1
an
PPI t INVt* INVt*1 INVt INVt 1 PPI t 1 PPI t INVt ( INVt 1 INVt* INVt*1 ) PPI t 1
us
cr
Replacement value of inventories: For firms using the first-in-first-out (FIFO) method, inventories are valued at current cost, thus the book value equals the replacement value for them. For firms using the last-in-first-out (LIFO) or any other method, inventories are valued at historic cost. While converting the book value to the replacement value, for the first year, the book value is taken equal to the replacement value. Following Salinger and Summers (1983) and Whited (1992), the following formulas are used for the following years:
if INVt* INVt*1
M
where INVt is the replacement value of LIFO inventories at time t and INVt* is their reported book value.
Ac ce p
te
d
Market Value of Long-Term Debt: The method suggested by Bernanke and Campbell (1988) and Whited (1992) is used on converting the book value of long-term debt to the replacement value. Since the COMPUSTAT database provides only limited information on maturities of debt, it is necessary to construct the maturity distribution of long-term debt from historical information on debt issues. Firstly, following Brainard, Shoven and Weiss (1980), it is assumed that all longterm debts mature in twenty years. For the first year, each individual firm's maturity distribution is set equal to the aggregate taken from Historical Statistics of the United States, series X 499509, p. 1005 for the years 1961-1970. I give equal weight to the maturity distribution for years 1971-80. Then, if Djt is debt due in j years at time t, LTDt is the reported value of long-term debt at time t, and DIt is the amount of debt issued at time t, the maturity distribution is updated as follows:
D20t DIt LTDt ( LTDt 1 D1,t 1 )
if LTDt ( LTDt 1 D1,t 1 ) 0
D20t DIt 0
if LTDt ( LTDt 1 D1,t 1 ) 0
D jt D j 1,t 1,
j = 1, …, 19.
If LTDt ( LTDt 1 D1,t 1 ) 0 , debt due in one to nineteen years is scaled down by the factor: LTDt . LTDt 1 D1,t 1
42 Page 42 of 60
The actual values of debt due in one to five year are available in COMPUSTAT. These values are replaced by the calculated values and the rest of the maturity distribution is rescaled in order to be consistent with total amount of long-term debt:
D ajt D*jt ,
j = 1, …, 5;
ip t
5 ( D jt D*jt ) , D ajt D jt 1 j 1 20 D jt j 6
cr
j = 6, …, 20;
us
where D ajt is the adjusted value of debt due in j years and D*jt , is the reported value of debt due in j years for j equal to one to five.
IEX t
M
NLTDt LTDt
an
The final modification adjusts the book value of total book value of debt to the reported interest expense consistent with that implied by assuming that the firm's interest expense at time t is the Baa rate at time t. The new value of total book value is scaled as follows:
,
20
Baa j 1
t j 20
D
a jt
te
d
where NLTDt is the scaled value of long-term debt at time t, Baat is the interest rate on grade Baa bonds at time t and IEXt is the book value of interest expense at time t. Then the new maturity distribution is set proportional to the old distribution.
Ac ce p
Market value of Equity: The value of common stock at the beginning of each year is estimated, following Salinger and Summers (1983), as the closing price of a share of stock for each company in year t-1 times the number of outstanding shares at t-1. The value of preferred stock is estimated by dividing preferred cash dividends by the Standard and Poor's preferred stock yield (taken from CITIBAS database).
43 Page 43 of 60
ip t cr
Table 1 - Average Values of Variables for Different Firm Classifications and Correlations between Fundamentals and Investment
Firms with Firms with low high number of number of employees employees
Firms with Firms with low high dividend dividend payout payout ratio ratio
1511.12
6943.51
27.71
3034.49
43.04
1880.46
8121.73
33.05
3322.61
51.23
0.13
0.34
0.13
0.12
0.14
13.40
48.95
0.86
25.24
0.66
0.15
4.24
0.21
0.07
0.22
0.68
3.86
0.81
0.56
0.90
0.39
0.65
0.41
0.35
0.30
4.11
0.22
0.04
0.18
0.21
Dividend to capital ratio (9) (10)
Firms with Firms with low high dividend dividend to capital to capital ratio ratio
Bond rating (11) (12) Firms without bond rating
Firms with bond rating
Total debt to capital ratio (13) (14) Firms with Firms with high debt low debt to capital to capital ratio ratio
KZ index (15) (16) Equitydependent firms (high constraint)
Not equitydependent firms (low constraint)
3005.50
965.70
2057.19
437.22
2592.75
232.01
4805.24
1214.08
1823.18
1246.19
1727.06
3289.48
1084.36
2228.98
498.36
2822.17
256.00
5261.34
1406.09
1922.43
1440.33
2052.53
0.12
0.15
0.11
0.14
0.11
0.13
0.12
0.14
0.12
0.12
0.13
25.25
11.28
14.58
4.61
21.21
3.08
38.07
16.07
9.80
10.49
14.54
0.07
0.24
0.05
0.23
0.06
0.17
0.08
0.21
0.08
0.19
0.22
0.47
1.04
0.34
0.91
0.47
0.69
0.67
0.76
0.61
0.81
0.57
0.41
0.35
0.46
0.30
0.48
0.29
0.38
0.39
0.60
0.16
0.54
0.25
0.40
0.11
0.51
-0.09
0.70
0.08
0.54
0.32
0.30
0.23
0.39
0.14
0.44
0.05
0.04
0.04
0.05
0.01
0.08
0.01
0.08
0.05
0.04
0.03
0.06
0.01
0.07
0.49
0.22
0.20
0.21
0.21
0.18
0.24
0.15
0.27
0.22
0.20
0.16
0.26
0.12
0.28
2.86
2.80
3.25
2.33
3.02
2.57
2.82
2.77
2.64
2.94
3.01
2.25
3.03
2.56
2.69
3.16
0.86
1.89
1.13
0.47
1.07
0.53
0.91
0.71
0.81
0.81
0.97
0.38
0.70
0.91
0.67
1.12
0.00 0.01 0.02 0.25
Correlations Whole sample 0.254 0.183 0.290 0.126
0.20
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.28
0.01
0.00
0.01
0.00
0.01
0.00
0.01
0.00
0.01
0.00
0.01
0.01
0.01
0.00
0.29
0.02
0.01
0.01
0.02
0.02
0.01
0.02
0.01
0.02
0.01
0.01
0.02
0.01
0.02
0.50
0.37
0.13
0.37
0.14
0.36
0.15
0.28
0.22
0.30
0.12
0.26
0.24
0.12
0.38
0.281 0.204 0.287 0.110
0.191 0.146 0.295 0.251
0.274 0.195 0.293 0.113
0.204 0.161 0.285 0.215
0.239 0.173 0.288 0.120
0.289 0.222 0.300 0.163
0.226 0.169 0.290 0.117
0.308 0.226 0.293 0.221
0.275 0.200 0.290 0.114
0.143 0.116 0.288 0.288
0.258 0.185 0.326 0.109
0.235 0.186 0.241 0.251
0.271 0.204 0.307 0.121
0.237 0.158 0.267 0.241
Ac
Corr(inv ,ait ) Corr(inv ,rait ) Corr(inv ,zit ) Corr(inv ,qit )
Firms with Firms with small large capital capital stock stock
ed
Fundamentals Profitability shocks (ait ) Profitability shocks (rait ) Mandated inv. rate (zit ) Log of Tobin's q
Dividend payout ratio (7) (8)
ce pt
Plant, Property, and Equipment (PPE, in millions of 1992 dollars) PPE (gross book value, in millions of dollars) Investment rate Employees (in thousands) Growth rate of real (net) sales Earnings retention rate Total Debt over PPE Dividends payout ratio Dividends over PPE Cash flow to PPE ratio Sales (net) to PPE ratio Working capital to PPE ratio
Standard deviation
Number of employees (5) (6)
us
Mean
Size of capital stock (3) (4)
M an
Whole sample (1) (2)
Note: Corr(.) is the correlation. inv is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q.
44 Page 44 of 60
ip t
Class 1: small capital stock
ait2
0.26 0.26 0.16 0.16 [17.5] [17.6] [12.3] [12.3] 0.20 0.26 [3.9] [5.3]
FV ait*FV Adj. R2 0.22 0.22 0.25 0.26 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.11 [3.88] Class 1 ait2 - Class 2 ait2 -0.07 [-2.39] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 0.11 0.12 [11.3] [11.9] [10.2] [11.1] 0.11 0.14 [4.4] [5.3]
rait*FV
(5)
(8)
0.24 0.24 [16.5] [16.6] 0.17 [3.4] 0.06 0.06 [5.5] [5.5]
0.23 [15]
0.12 0.13 [10.4] [10.9] 0.09 [3.7] 0.04 0.03 [2.5] [2.3]
0.12 [9.9]
0.38 0.06 [14] [5.6] 0.07 [2.8] 0.23 0.23 0.10 0.23 Column (6) & Column (10) 0.09 [3.66] -0.10 [-5.29] -0.07 [-4.79]
0.38 0.02 [14] [1.7] 0.05 [2.1] 0.19 0.19 0.10 0.19 Column (6) & Column (10) 0.03 [2] -0.04 [-4.99] -0.07 [-4.5]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.18 0.18 0.23 0.24 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.01 [2.74] Class 1 rait2 -Class 2 rait2 -0.03 [-0.4] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.15 0.15 0.27 [11.9] [11.8] [13] 0.27 [5.5] 0.14 0.14 0.16 0.22 [7.5] [7.7] [8.4] [10.3] -0.60 [-7.3] 0.26 0.27 0.23 0.28
0.09 0.10 0.09 [8.3] [9.2] [5.5] 0.13 [5] 0.11 0.10 0.16 0.11 [5.4] [5.1] [8.4] [5.4] 0.01 [0.2] 0.24 0.24 0.23 0.24
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: large capital stock
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.22 0.22 0.11 0.11 0.11 0.10 0.23 0.23 0.16 0.13 0.13 0.11 [13.7] [13.8] [5.2] [8.6] [8.6] [4.5] [15.6] [15.7] [8.6] [10.3] [10.3] [5.9] 0.19 0.25 0.19 0.23 [3.7] [5.1] [3.8] [4.8] 0.02 0.02 0.16 0.03 0.06 0.06 0.07 0.06 0.04 0.04 0.19 0.05 0.13 0.13 0.15 0.13 [8] [7.9] [32.8] [9] [13.7] [13.7] [17.3] [13.5] [10.2] [10.2] [19.6] [10.8] [11.4] [11.2] [13.2] [11.1] 0.03 0.00 0.06 0.04 [6.7] [0.3] [7] [1.6] 0.24 0.24 0.31 0.25 0.30 0.31 0.29 0.30 0.25 0.25 0.15 0.26 0.29 0.29 0.26 0.29 Column (14) & Column (18) Column (22) & Column (26) 0.11 [1.76] 0.10 [1.92] -0.06 [-5.29] -0.04 [-4.95] -0.04 [-4.79] -0.08 [-1.39]
M an
Financial Variables
cr
Table 2 -Response of investment to alternative fundamentals and financial variables (when firms are grouped by the size of their capital stocks)
0.07 0.08 0.11 [6.2] [6.7] [6.4] 0.07 [3.2] 0.03 0.03 0.16 0.04 [10.1] [9.6] [32.8] [9.6] 0.02 [2.9] 0.21 0.21 0.31 0.21 Column (14) & Column (18) 0.03 [1.76] -0.03 [-4.38] -0.03 [-5.5]
0.04 0.05 0.09 [3.8] [4.6] [5] 0.10 [4] 0.07 0.06 0.07 0.07 [14] [13.6] [17.3] [14.3] -0.02 [-3.3] 0.28 0.29 0.29 0.29
0.09 [8.6]
0.10 0.12 [9] [9.3] 0.08 [3.3] 0.05 0.05 0.19 0.06 [11.2] [10.7] [19.6] [11.6] 0.03 [3.9] 0.21 0.22 0.15 0.22 Column (22) & Column (26) 0.00 [1.44] -0.04 [-3.59] -0.08 [-0.2]
0.09 0.10 0.12 [8.6] [9.3] [8] 0.11 [4.3] 0.14 0.13 0.15 0.14 [12] [11.5] [13.2] [11.9] -0.06 [-2.9] 0.27 0.28 0.26 0.27
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
45 Page 45 of 60
ip t
Class 1: small capital stock
zit2
0.21 0.20 0.19 0.19 [19.3] [18.8] [20.1] [19.1] 0.08 0.05 [4] [2.6]
FV zit*FV Adj. R2 0.22 0.22 0.31 0.31 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.02 [1.8] Class 1 zit2 -Class 2 zit2 0.03 [0.75] Class 1 FV - Class 2 FV qit qit2
0.29 [7.8]
0.28 0.12 0.11 [5.4] [10.9] [8.5] 0.01 0.04 [0.3] [2.5]
qit*FV
(5) 0.21 [18.4]
(8)
0.20 [18] 0.08 [4] -0.02 -0.02 [-1.3] [-1.3]
0.20 [18.1]
0.24 0.24 [6.7] [4.8] 0.01 [0.2] 0.54 0.54 [15.6] [15.6]
0.20 [5.5]
0.38 -0.03 [14] [-2] 0.03 [1.6] 0.23 0.23 0.10 0.23 Column (6) & Column (10) 0.03 [2.67] 0.03 [1.86] -0.09 [-2.58]
0.38 0.35 [14] [5.8] 0.18 [3.8] 0.14 0.14 0.10 0.14 Column (6) & Column (10) 0.14 [2.35] -0.02 [-2.01] 0.38 [0.79]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.05 0.26 0.26 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.18 [1.87] Class 1 qit2 -Class 2 qit2 -0.03 [1.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.18 0.18 0.20 [18.8] [17.7] [16.2] 0.05 [2.8] 0.07 0.07 0.16 0.11 [3.7] [3.8] [8.4] [3.9] -0.10 [-2] 0.32 0.32 0.23 0.32
0.11 0.10 0.16 [9.7] [7.8] [11] 0.03 [1.8] 0.15 0.15 0.16 0.30 [7.9] [7.8] [8.4] [9.3] -0.27 [-5.7] 0.28 0.28 0.23 0.29
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: large capital stock
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.20 0.20 0.19 0.15 0.14 0.14 0.18 0.17 0.19 0.17 0.16 0.12 [14.1] [14.1] [12.7] [12.1] [11.4] [8.3] [15.2] [15] [14.7] [16.7] [16] [9.1] 0.08 0.05 0.08 0.04 [3.9] [2.7] [3.6] [2.1] 0.00 0.00 0.16 0.00 0.03 0.03 0.07 0.02 0.03 0.03 0.19 0.04 0.09 0.08 0.15 0.06 [1.1] [0.7] [32.8] [0.3] [5.2] [5.3] [17.3] [3.1] [5.8] [5.6] [19.6] [5.9] [7.4] [7.2] [13.2] [5.2] 0.01 0.01 0.02 0.09 [2.3] [1.1] [2.5] [1.4] 0.22 0.22 0.31 0.22 0.32 0.32 0.29 0.32 0.23 0.23 0.15 0.23 0.33 0.33 0.26 0.33 Column (14) & Column (18) Column (22) & Column (26) 0.05 [2.16] 0.01 [2.19] 0.03 [1.76] 0.04 [1.52] -0.03 [-6.95] -0.05 [-2.12]
M an
Financial Variables
cr
Table 2 (cont'd) -Response of investment to alternative fundamentals and financial variables (when firms are grouped by the size of their capital stocks)
0.15 [4.9]
0.13 -0.06 0.09 0.08 0.07 0.18 0.17 0.18 [3] [-1.6] [8.9] [7] [4.1] [5.1] [3.4] [4.7] 0.02 0.03 0.01 [0.7] [2.1] [0.5] 0.18 0.18 0.16 0.11 0.06 0.06 0.07 0.05 0.21 0.21 0.19 0.21 [32.9] [32.9] [32.8] [11.2] [15.3] [15.3] [17.3] [13.5] [18.7] [18.7] [19.6] [9.5] 0.07 0.01 0.01 [9.6] [1.5] [1.8] 0.35 0.34 0.31 0.37 0.33 0.33 0.29 0.33 0.17 0.17 0.15 0.17 Column (14) & Column (18) Column (22) & Column (26) 0.05 [1.85] 0.07 [1.88] -0.01 [-1.85] -0.02 [-2.39] 0.12 [6.98] 0.11 [1.71]
0.11 0.09 0.10 [9.5] [7.3] [7.3] 0.04 [2.4] 0.11 0.11 0.15 0.11 [10] [10] [13.2] [7] 0.00 [0.1] 0.29 0.29 0.26 0.29
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
46 Page 46 of 60
ip t
Class 1: low number of employees
ait2
0.26 0.26 0.17 0.17 [17.1] [17.1] [13.2] [13.2] 0.18 0.30 [3.4] [6.2]
FV ait*FV Adj. R2 0.22 0.22 0.24 0.25 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.09 [2.24] Class 1 ait2 - Class 2 ait2 -0.13 [-2.96] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 [10.9] [11.5] 0.10 [4.4]
0.11 [11]
0.13 [12] 0.15 [5.5]
rait*FV
(8)
0.24 0.24 0.23 [16.2] [16.3] [14.9] 0.15 [2.9] 0.05 0.05 0.36 0.05 [4.7] [4.7] [13.4] [4.7] 0.06 [2.5] 0.23 0.23 0.09 0.23 Column (6) & Column (10) 0.09 [2.19] -0.17 [-2.06] -0.16 [-5.34] 0.12 [10.5]
0.13 0.11 [11] [9.8] 0.09 [3.8] 0.02 0.02 0.36 0.01 [1.6] [1.4] [13.4] [0.9] 0.05 [2.4] 0.19 0.20 0.09 0.20 Column (6) & Column (10) 0.03 [2.19] -0.05 [-4.59] -0.16 [-5.34]
ce pt
FV
(5)
Class 1 (6) (7)
Adj. R2 0.18 0.19 0.22 0.23 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.00 [1.74] Class 1 rait2 -Class 2 rait2 -0.04 [-1.84] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.15 0.15 0.21 [12.2] [12.1] [10] 0.32 [6.6] 0.21 0.21 0.24 0.23 [10.9] [11.2] [12.3] [11.5] -0.26 [-3.4] 0.27 0.28 0.23 0.27
0.21 0.21 0.11 0.12 0.12 0.10 0.23 0.23 0.15 0.14 0.14 0.10 [13.5] [13.6] [5.1] [9.2] [9.2] [4.5] [15.1] [15.2] [8.5] [11.3] [11.2] [5.7] 0.16 0.29 0.17 0.29 [3.2] [6.1] [3.3] [6] 0.02 0.02 0.17 0.03 0.05 0.05 0.06 0.05 0.04 0.04 0.19 0.04 0.13 0.13 0.14 0.13 [7.8] [7.7] [33.6] [8.9] [13.5] [13.5] [16.9] [12.9] [9.8] [9.7] [19.5] [10.2] [12.8] [12.8] [14.1] [12.9] 0.03 0.01 0.06 0.07 [6.9] [1.1] [7.1] [1.3] 0.24 0.24 0.32 0.25 0.29 0.30 0.27 0.29 0.25 0.25 0.15 0.26 0.28 0.29 0.25 0.28 Column (14) & Column (18) Column (22) & Column (26) 0.10 [2.65] 0.09 [2.99] -0.13 [-2.01] -0.12 [-2.01] -0.03 [-2.39] -0.09 [-8.82]
0.08 0.09 0.13 [7.5] [8.5] [7.6] 0.14 [5.1] 0.18 0.18 0.24 0.20 [8.7] [8.5] [12.3] [9.4] -0.23 [-3.6] 0.24 0.25 0.23 0.24
0.07 0.08 0.10 0.05 0.06 0.11 0.09 0.10 0.12 0.09 0.10 0.12 [6.1] [6.6] [6.1] [4.4] [5.2] [6.7] [8.3] [8.7] [9.2] [8.9] [9.7] [7.9] 0.07 0.11 0.08 0.12 [3.2] [4.3] [3.3] [4.6] 0.03 0.03 0.17 0.04 0.05 0.05 0.06 0.06 0.05 0.05 0.19 0.06 0.13 0.12 0.14 0.13 [9.9] [9.4] [33.6] [9.2] [13.4] [13] [16.9] [14.2] [10.9] [10.5] [19.5] [11.5] [12.3] [11.9] [14.1] [12.5] 0.01 -0.03 0.03 -0.04 [2.5] [-5] [4.1] [-2.5] 0.21 0.21 0.32 0.21 0.27 0.27 0.27 0.27 0.22 0.22 0.15 0.22 0.26 0.27 0.25 0.26 Column (14) & Column (18) Column (22) & Column (26) 0.02 [2.65] -0.01 [-0.99] -0.04 [-4.2] -0.05 [-4.09] -0.02 [-2.39] -0.08 [-8.82]
ed
ait
FV=CF_K
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high number of employees
M an
Financial Variables
cr
Table 3 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the number of employees)
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t -statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
47 Page 47 of 60
ip t
Class 1: low number of employees
zit2
0.20 0.20 0.19 0.19 [19.2] [18.8] [20.3] [19.1] 0.07 0.07 [3.5] [3.5]
FV zit*FV Adj. R2 0.23 0.23 0.30 0.31 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.01 [1.76] Class 1 zit2 -Class 2 zit2 0.00 [0.21] Class 1 FV - Class 2 FV qit qit2
0.30 [8.1]
0.30 0.11 0.09 [5.7] [10.2] [7.8] 0.00 0.05 [0] [3]
qit*FV
(5)
(8)
0.21 0.20 0.20 [18.7] [18.3] [18.4] 0.07 [3.5] -0.02 -0.02 0.36 -0.03 [-1.7] [-1.6] [13.4] [-2.2] 0.03 [2.5] 0.23 0.23 0.09 8.00 Column (6) & Column (10) 0.03 [1.76] 0.00 [2.24] -0.13 [-4.72] 0.25 [7]
0.25 0.21 [5] [5.7] 0.00 [0] 0.49 0.49 0.36 0.28 [14.6] [14.6] [13.4] [4.6] 0.21 [4.2] 0.13 0.13 0.09 0.13 Column (6) & Column (10) 0.17 [1.92] -0.02 [-1.82] 0.16 [0.87]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.05 0.24 0.24 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.21 [3.87] Class 1 qit2 -Class 2 qit2 -0.05 [-1.73] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.18 0.17 0.18 [17.4] [16.4] [13.9] 0.07 [3.5] 0.11 0.11 0.24 0.13 [5.6] [5.7] [12.3] [4.6] -0.05 [-0.9] 0.31 0.31 0.23 0.31
0.09 0.08 0.09 [8.2] [6.8] [6] 0.02 [1.2] 0.33 0.33 0.24 0.34 [14.8] [14.5] [12.3] [12.2] -0.01 [-0.3] 0.30 0.30 0.23 0.30
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high number of employees
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.19 0.19 0.19 0.16 0.15 0.15 0.18 0.17 0.19 0.17 0.17 0.14 [14.2] [14.1] [13.1] [12.9] [12.4] [9.2] [15.4] [15.1] [14.7] [16.7] [16] [10.4] 0.07 0.06 0.07 0.06 [3.4] [3.2] [3.2] [2.9] 0.00 0.00 0.17 0.00 0.02 0.02 0.06 0.02 0.03 0.03 0.19 0.04 0.08 0.08 0.14 0.07 [1.2] [0.9] [33.6] [0.5] [4.6] [4.4] [16.9] [2.2] [5.7] [5.5] [19.5] [5.6] [8] [7.7] [14.1] [5.5] 0.01 0.01 0.02 0.06 [2.1] [1] [2.2] [1.2] 0.23 0.23 0.32 0.22 0.31 0.31 0.27 0.31 0.23 0.24 0.15 0.24 0.32 0.32 0.25 0.32 Column (14) & Column (18) Column (22) & Column (26) 0.04 [1.74] 0.01 [1.94] 0.01 [1.62] 0.01 [1.3] -0.02 [-3.01] -0.05 [-7.31]
M an
Financial Variables
cr
Table 3 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the number of employees)
0.15 [4.8]
0.13 -0.04 0.09 0.08 0.06 0.19 0.18 0.19 [3] [-1.2] [8.6] [6.6] [3.7] [5.3] [3.6] [4.9] 0.02 0.04 0.01 [0.6] [2.5] [0.3] 0.19 0.19 0.17 0.11 0.05 0.05 0.06 0.05 0.21 0.21 0.19 0.21 [33.2] [33.2] [33.6] [11] [16.3] [16.2] [16.9] [14.6] [18.4] [18.4] [19.5] [9.2] 0.07 0.01 0.00 [8.9] [1.2] [2.1] 0.35 0.35 0.32 0.37 0.31 0.31 0.27 0.31 0.17 0.17 0.15 0.17 Column (14) & Column (18) Column (22) & Column (26) 0.06 [2.21] 0.10 [1.76] -0.02 [-1.61] -0.03 [-2.01] 0.13 [1.48] 0.09 [0.21]
0.09 0.08 0.09 [8.6] [6.5] [6.3] 0.04 [2.7] 0.13 0.13 0.14 0.12 [12] [11.9] [14.1] [9.9] 0.01 [0.3] 0.28 0.28 0.25 0.28
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
48 Page 48 of 60
ip t
Class 1: low dividend payout ratio
ait2
0.24 0.24 0.19 0.19 [15.3] [15.4] [18.2] [18.1] 0.17 0.32 [3.1] [7.6]
FV ait*FV Adj. R2 0.22 0.22 0.25 0.26 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.06 [6.29] Class 1 ait2 - Class 2 ait2 -0.15 [-3.01] Class 1 FV - Class 2 FV rait rait2
0.12 [10.4]
0.13 0.13 0.13 [11] [14.8] [15.6] 0.11 0.12 [4.3] [5.7]
rait*FV
(5)
(8)
Adj. R2 0.19 0.19 0.22 0.23 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.00 [4.4] Class 1 rait2 -Class 2 rait2 -0.01 [-1.66] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.23 0.23 0.22 [14.3] [14.4] [13.2] 0.14 [2.7] 0.07 0.07 0.39 0.06 [5.2] [5.2] [13.9] [4.4] 0.06 [2.6] 0.23 0.23 0.10 0.23 Column (6) & Column (10) 0.05 [4.03] -0.17 [-2.58] -0.05 [-2.81]
0.18 0.18 0.15 [17.6] [17.5] [12.7] 0.32 [7.7] 0.12 0.12 0.14 0.16 [9.4] [9.4] [10.5] [10.5] 0.09 [1.7] 0.27 0.29 0.18 0.27
0.19 0.19 0.08 0.17 0.17 0.09 0.20 0.20 0.12 0.19 0.19 0.17 [10.8] [10.9] [3] [16] [15.9] [5.7] [12.7] [12.8] [6] [17.8] [17.7] [13.3] 0.16 0.30 0.16 0.31 [2.9] [7.4] [3] [7.4] 0.03 0.03 0.19 0.02 0.03 0.03 0.03 0.03 0.06 0.06 0.23 0.05 0.02 0.02 0.03 0.03 [8.6] [8.5] [36.6] [6.5] [11.5] [11.3] [14.5] [13.6] [11] [10.9] [21.4] [9.2] [5.3] [5.1] [5.9] [6] 0.04 0.02 0.10 0.02 [5.7] [1.2] [6.9] [1.3] 0.24 0.24 0.36 0.25 0.28 0.30 0.21 0.29 0.25 0.26 0.18 0.27 0.25 0.27 0.16 0.25 Column (14) & Column (18) Column (22) & Column (26) 0.02 [2.13] 0.02 [1.83] -0.15 [-2.33] -0.15 [-2.46] 0.01 [1.95] 0.03 [4.89]
0.12 0.13 0.11 [9.7] [10.2] [8.9] 0.09 [3.7] 0.04 0.03 0.39 0.04 [2.4] [2.2] [13.9] [2.3] 0.03 [2.4] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.01 [1.69] -0.02 [-1.75] -0.05 [-5.04]
0.10 0.11 0.11 [11.4] [12.2] [10.9] 0.12 [5.5] 0.08 0.08 0.14 0.09 [5.7] [5.5] [10.5] [4.2] -0.01 [-0.3] 0.23 0.24 0.18 0.23
0.06 0.07 0.09 [4.8] [5.3] [4.9] 0.07 [2.6] 0.04 0.04 0.19 0.05 [11.1] [10.5] [36.6] [10.4] 0.01 [2.1] 0.22 0.23 0.36 0.22 Column (14) & Column (18) -0.03 [-0.77] -0.04 [-1.77] 0.02 [6.22]
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend payout ratio
M an
Financial Variables
cr
Table 4 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend payout ratio)
0.09 0.10 0.13 [10] [10.8] [10.3] 0.11 [5.3] 0.02 0.02 0.03 0.03 [9.6] [9.3] [14.5] [10.1] -0.02 [-4.6] 0.24 0.25 0.21 0.25
0.09 [7.6]
0.10 0.12 0.12 0.13 0.14 [8] [8.6] [13.7] [14.5] [12.2] 0.07 0.12 [2.8] [5.7] 0.07 0.07 0.23 0.08 0.02 0.02 0.03 0.03 [12.3] [11.9] [21.4] [12.9] [3.3] [3.2] [5.9] [4.1] 0.03 -0.02 [3.9] [-2.4] 0.23 0.23 0.18 0.24 0.22 0.23 0.16 0.22 Column (22) & Column (26) -0.03 [-0.61] -0.05 [-1.96] 0.05 [5.91]
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
49 Page 49 of 60
ip t
Class 1: low dividend payout ratio
zit2
0.22 0.21 0.17 0.17 [19.6] [18.9] [20.6] [20.4] 0.05 0.14 [2.3] [6.9]
FV zit*FV Adj. R2 0.25 0.25 0.28 0.29 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.04 [4.81] Class 1 zit2 -Class 2 zit2 -0.09 [-2.6] Class 1 FV - Class 2 FV qit qit2
0.33 [8.7]
0.35 0.07 0.06 [6.5] [7.6] [5.2] -0.01 0.04 [-0.4] [3.4]
qit*FV
(5)
(8)
Adj. R2 0.06 0.06 0.17 0.17 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.29 [3.87] Class 1 qit2 -Class 2 qit2 -0.05 [-2.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
0.16 [18]
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.21 0.21 0.21 [18.8] [18.2] [18.3] 0.05 [2.3] -0.01 -0.01 0.39 -0.01 [-0.4] [-0.4] [13.9] [-0.8] 0.02 [1.8] 0.25 0.25 0.10 0.25 Column (6) & Column (10) 0.05 [2.03] -0.09 [-2.74] -0.03 [-4.69]
0.16 0.16 [18] [17.4] 0.14 [6.8] 0.03 0.02 0.14 0.02 [1.9] [1.5] [10.5] [0.7] 0.02 [0.7] 0.28 0.29 0.18 0.28
0.20 0.20 0.21 0.15 0.15 0.15 0.18 0.18 0.20 0.17 0.17 0.16 [13.7] [13.4] [12.1] [15.4] [15.6] [13.2] [14.9] [14.5] [14.5] [19.5] [19.4] [16.2] 0.05 0.13 0.04 0.14 [2.2] [6.6] [2] [7] 0.01 0.01 0.19 0.01 0.01 0.01 0.03 0.00 0.05 0.05 0.23 0.06 0.00 0.00 0.03 -0.01 [1.2] [1.1] [36.6] [0.9] [3.4] [2.7] [14.5] [0.6] [7.4] [7.4] [21.4] [7.5] [-0.2] [-0.7] [5.9] [-2] 0.00 0.01 0.03 -0.02 [1.2] [1.2] [2.8] [-2.4] 0.25 0.25 0.36 0.24 0.28 0.29 0.21 0.28 0.26 0.26 0.18 0.26 0.28 0.29 0.16 0.28 Column (14) & Column (18) Column (22) & Column (26) 0.05 [2.64] 0.01 [2.05] -0.09 [-2.54] -0.10 [-2.23] 0.00 [2.75] 0.05 [6.73]
0.27 0.28 [7.4] [5.5] -0.01 [-0.3] 0.59 0.59 0.39 [16.1] [16.1] [13.9]
0.07 0.05 0.09 [6.9] [4.9] [8.7] 0.03 [2.8] 0.13 0.13 0.14 0.25 [9.2] [9] [10.5] [10.4] -0.10 [-6.2] 0.20 0.20 0.18 0.21
0.15 0.12 -0.35 0.06 0.05 0.11 0.20 0.18 0.16 [4.8] [2.8] [-10.1] [6.8] [4.8] [9.4] [5.6] [3.6] [4.2] 0.02 0.03 0.02 [0.9] [2.7] [0.6] 0.22 0.22 0.19 0.05 0.03 0.03 0.03 0.05 0.27 0.27 0.23 0.21 [37.3] [37.3] [36.6] [5.6] [12.9] [12.7] [14.5] [13.6] [21.2] [21.2] [21.4] [8.6] 0.19 -0.02 0.05 [23.7] [-6.6] [2.6] 0.41 0.41 0.36 0.52 0.22 0.22 0.21 0.24 0.21 0.21 0.18 0.21 Column (14) & Column (18) Column (22) & Column (26) 0.07 [2.48] 0.13 [2.78] -0.01 [-1.97] -0.02 [-3.15] 0.19 [13.77] 0.25 [7.2]
0.19 [5]
0.18 [3] 0.46 [8] 0.15 0.15 0.10 0.17 Column (6) & Column (10) 0.23 [3.3] -0.04 [-2.57] 0.46 [10.44]
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend payout ratio
M an
Financial Variables
cr
Table 4 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend payout ratio)
0.07 [7.1]
0.02 [4.6]
0.18
0.06 0.07 [5] [6.8] 0.04 [2.9] 0.02 0.03 0.03 [4.3] [5.9] [2.9] 0.00 [-0.5] 0.18 0.16 0.18
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
50 Page 50 of 60
ip t
Class 1: low dividend to capital ratio
ait2
0.23 0.23 0.21 0.21 [14.6] [14.6] [19.2] [19.6] 0.13 0.41 [2.5] [9]
FV ait*FV Adj. R2 0.21 0.21 0.27 0.29 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.01 [5.07] Class 1 ait2 - Class 2 ait2 -0.28 [-2.01] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 0.13 0.14 [10.2] [10.8] [14.6] [15.5] 0.12 0.12 [4.6] [5.3]
rait*FV
(5)
(8)
Adj. R2 0.19 0.19 0.23 0.24 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait -0.01 [-1.52] Class 1 rait2 -Class 2 rait2 0.00 [0.7] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.23 0.23 0.21 [13.9] [13.9] [13.1] 0.23 [2.1] 0.06 0.06 0.41 0.05 [4.2] [4.3] [14.1] [3.7] 0.04 [2.5] 0.22 0.22 0.10 0.22 Column (6) & Column (10) 0.06 [2.79] 0.08 [4.51] -0.08 [-7.75]
0.17 0.17 0.15 [17.8] [18.2] [12] 0.15 [8.6] 0.15 0.14 0.16 0.17 [12.4] [12.1] [13.1] [13.7] 0.11 [1.2] 0.31 0.33 0.23 0.32
0.12 0.13 0.12 [10.1] [10.6] [9.3] 0.10 [4] 0.02 0.02 0.41 0.02 [1.4] [1.1] [14.1] [1.2] 0.03 [1.3] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.03 [0.61] -0.01 [-0.7] -0.10 [-2.92]
0.09 0.10 0.10 [9.5] [10.4] [9.6] 0.11 [4.9] 0.12 0.12 0.16 0.14 [9] [8.8] [13.1] [8.1] -0.06 [-1.9] 0.25 0.26 0.23 0.25
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend capital ratio
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.19 0.20 0.07 0.15 0.15 0.09 0.20 0.20 0.10 0.17 0.17 0.17 [10.5] [10.6] [2.8] [16] [16.4] [5.6] [12.3] [12.3] [5.2] [18.3] [18.6] [12.8] 0.25 0.13 0.25 0.13 [2.4] [8.5] [2.6] [8.4] 0.03 0.03 0.21 0.03 0.03 0.03 0.04 0.03 0.06 0.06 0.26 0.05 0.04 0.04 0.05 0.04 [8.1] [8.1] [38.2] [6.6] [13.4] [13.1] [16.6] [15.5] [10.1] [10.1] [22.7] [9.6] [10.8] [10.4] [12.2] [11.5] 0.04 0.02 0.11 0.03 [5.7] [0.9] [7.3] [0.5] 0.23 0.23 0.38 0.24 0.32 0.34 0.25 0.33 0.24 0.24 0.20 0.26 0.30 0.32 0.22 0.31 Column (14) & Column (18) Column (22) & Column (26) 0.05 [2.77] 0.03 [2.45] 0.11 [4.12] 0.12 [4.07] 0.00 [4.59] 0.02 [1.63]
M an
Financial Variables
cr
Table 5 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend to capital ratio)
0.07 0.07 0.09 [5.2] [5.7] [5.2] 0.07 [3] 0.04 0.04 0.21 0.05 [10.4] [9.7] [38.2] [10] 0.01 [2.2] 0.22 0.22 0.38 0.22 Column (14) & Column (18) -0.02 [-0.99] -0.03 [-1.54] 0.01 [4.57]
0.08 0.09 0.13 [8.3] [9.1] [9.3] 0.10 [4.7] 0.03 0.03 0.04 0.04 [12] [11.8] [16.6] [12] -0.02 [-4.9] 0.27 0.28 0.25 0.28
0.09 [7.8]
0.10 0.12 0.11 0.12 0.14 [8.2] [8.7] [12.3] [13.1] [11.8] 0.08 0.11 [3.1] [5] 0.07 0.07 0.26 0.08 0.04 0.04 0.05 0.05 [11.4] [10.9] [22.7] [12.1] [9.3] [9.1] [12.2] [9.6] 0.04 -0.03 [4] [-3.7] 0.23 0.23 0.20 0.23 0.25 0.26 0.22 0.26 Column (22) & Column (26) -0.03 [-0.73] -0.03 [-1.21] 0.03 [2.53]
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
51 Page 51 of 60
ip t
Class 1: low dividend to capital ratio
zit2
0.22 0.21 0.17 0.16 [19.9] [19.1] [19.8] [19.4] 0.06 0.09 [3] [4.6]
FV zit*FV Adj. R2 0.25 0.25 0.28 0.28 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.05 [2.43] Class 1 zit2 -Class 2 zit2 -0.02 [-1.78] Class 1 FV - Class 2 FV qit qit2
0.33 [8.5]
0.33 0.08 0.07 [6.1] [8.4] [6.2] 0.00 0.03 [0] [2.4]
qit*FV
(5)
(8)
Adj. R2 0.06 0.06 0.23 0.23 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.26 [1.87] Class 1 qit2 -Class 2 qit2 -0.03 [-2.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.21 0.21 0.21 [19.2] [18.5] [18.7] 0.06 [3] -0.01 0.00 0.41 -0.01 [-0.4] [-0.3] [14.1] [-0.9] 0.02 [2.1] 0.26 0.26 0.10 0.26 Column (6) & Column (10) 0.05 [1.71] -0.02 [-2.4] -0.03 [-3.81]
0.16 0.16 0.16 [16.7] [16.4] [16] 0.08 [4.5] 0.03 0.03 0.16 0.03 [2.2] [2] [13.1] [1.4] 0.00 [0.2] 0.28 0.28 0.23 0.28
0.27 0.28 0.16 [7.4] [5.5] [4.4] -0.01 [-0.2] 0.62 0.62 0.41 0.10 [16.7] [16.7] [14.1] [1.6] 0.61 [10] 0.16 0.16 0.10 0.19 Column (6) & Column (10) 0.22 [3] -0.03 [-3.13] 0.50 [10.93]
0.07 0.06 0.09 [7.3] [5.5] [8.7] 0.02 [1.9] 0.13 0.13 0.16 0.23 [9.5] [9.4] [13.1] [9.7] -0.08 [-5.3] 0.22 0.22 0.23 0.23
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend capital ratio
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.20 0.20 0.21 0.15 0.14 0.14 0.18 0.18 0.20 0.17 0.16 0.15 [13.9] [13.7] [12.5] [14.1] [13.9] [11.8] [15.3] [14.8] [15] [18.3] [18] [15] 0.06 0.08 0.06 0.09 [2.9] [4.4] [2.7] [4.6] 0.01 0.01 0.21 0.01 0.01 0.01 0.04 0.00 0.05 0.05 0.26 0.07 0.00 0.00 0.05 -0.01 [1.5] [1.2] [38.2] [1.4] [3.6] [3.4] [16.6] [0.2] [8] [7.9] [22.7] [8.1] [0.6] [0.2] [12.2] [-1.5] 0.00 0.01 0.03 0.02 [1.6] [0.8] [3.1] [1.4] 0.25 0.25 0.38 0.25 0.28 0.28 0.25 0.28 0.27 0.27 0.20 0.27 0.27 0.28 0.22 0.28 Column (14) & Column (18) Column (22) & Column (26) 0.06 [2.04] 0.01 [1.69] -0.02 [-2] -0.03 [-2.21] 0.00 [2.08] 0.05 [10.4]
M an
Financial Variables
cr
Table 5 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend to capital ratio)
0.14 0.11 -0.37 0.07 0.06 0.11 0.21 0.19 0.16 [4.6] [2.6] [-10.8] [7.4] [5.6] [9.2] [5.9] [3.9] [4.3] 0.03 0.02 0.01 [1.1] [1.8] [0.5] 0.23 0.23 0.21 0.05 0.03 0.03 0.04 0.05 0.29 0.29 0.26 0.22 [38.8] [38.8] [38.2] [6.1] [13.7] [13.6] [16.6] [13.2] [22.3] [22.3] [22.7] [8.7] 0.20 -0.01 0.06 [24.3] [-5.4] [3] 0.42 0.42 0.38 0.54 0.25 0.25 0.25 0.26 0.22 0.22 0.20 0.22 Column (14) & Column (18) Column (22) & Column (26) 0.05 [0.68] 0.13 [2.21] 0.01 [2.13] -0.01 [-3.25] 0.20 [13.62] 0.26 [6.68]
0.08 0.06 0.08 [7.4] [5.5] [7.1] 0.02 [2] 0.03 0.03 0.05 0.03 [5.8] [5.6] [12.2] [3.5] 0.00 [-0.4] 0.20 0.21 0.22 0.20
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
52 Page 52 of 60
ip t
Class 1: firms without debt rating
ait2
0.25 0.25 0.25 0.13 [20.7] [20.8] [16.1] [7.6] 0.20 0.27 [4.8] [3.9]
FV ait*FV Adj. R2 0.23 0.23 0.19 0.24 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.12 [2.29] Class 1 ait2 - Class 2 ait2 -0.07 [-2.4] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 0.13 0.11 [13.6] [14.4] [10.8] [7.8] 0.11 0.18 [5.5] [4.9]
rait*FV
Adj. R2 0.19 0.19 0.15 0.23 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.02 [2.55] Class 1 rait2 -Class 2 rait2 -0.07 [-2.06] Class 1 FV - Class 2 FV
FV=Sales_K
Class 1 (13) (14) (15) (16)
0.23 0.23 0.22 [19.7] [19.8] [17.9] 0.18 [4.4] 0.07 0.07 0.34 0.06 [6.5] [6.5] [15.7] [6.5] 0.05 [2.4] 0.24 0.24 0.10 0.24 Column (6) & Column (10) 0.14 [1.79] -0.06 [-4.84] -0.40 [-6.76]
0.23 0.09 0.18 [15.5] [5.7] [5.7] 0.24 [3.6] 0.06 0.46 0.52 0.48 [4.2] [11.3] [12.8] [11.7] -0.41 [-3.3] 0.20 0.30 0.28 0.30
0.20 0.20 0.10 [16] [16.1] [5.5] 0.19 [4.5] 0.03 0.03 0.16 0.03 [10.7] [10.5] [38.7] [11.7] 0.03 [8.2] 0.25 0.25 0.31 0.26 Column (14) & Column (18) 0.11 [2.28] -0.08 [-4.47] -0.02 [-2.39]
0.22 0.22 0.15 0.20 [18.4] [18.4] [10.5] [12.4] 0.19 [4.7] 0.05 0.05 0.19 0.05 0.10 [13.2] [13.1] [23.7] [13.7] [11] 0.06 [7.7] 0.26 0.27 0.16 0.27 0.22 Column (22) & Column (26) 0.10 [2.45] -0.07 [-4.39] 0.03 [6.92]
0.12 [6.9] 0.27 [3.8] 0.02 [1.1]
0.12 0.13 0.11 [12.6] [13.3] [11.6] 0.09 [4.8] 0.04 0.03 0.34 0.03 [3] [2.7] [15.7] [2.3] 0.05 [2.4] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.09 [2.64] -0.06 [-4.42] -0.42 [-7.54]
0.14 0.04 0.09 [11.2] [2.5] [4] 0.15 [4.1] 0.00 0.46 0.52 0.50 [0.2] [9.6] [12.8] [10.5] -0.35 [-3.8] 0.17 0.28 0.28 0.28
0.07 0.08 0.11 0.07 0.06 0.09 0.09 0.10 0.13 0.09 [7.2] [7.9] [7.8] [5.3] [3.9] [4.1] [10.5] [11.1] [12.1] [7.5] 0.08 0.15 0.08 [3.9] [4.2] [4] 0.03 0.03 0.16 0.04 0.04 0.05 0.06 0.06 0.06 0.06 0.19 0.07 0.13 [12.8] [12.2] [38.7] [12.5] [10.7] [8.5] [11.3] [9.3] [14.2] [13.7] [23.7] [15.5] [13.5] 0.02 -0.02 0.04 [3.9] [-2.8] [2.2] 0.22 0.22 0.31 0.22 0.19 0.27 0.27 0.26 0.23 0.23 0.16 0.24 0.21 Column (14) & Column (18) Column (22) & Column (26) 0.02 [1.7] -0.01 [-1.01] -0.08 [-3.56] -0.12 [-2.23] -0.02 [-2.91] 0.03 [4.62]
0.11 [7.3] 0.19 [5] 0.02 [1.2]
(5)
(8)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K
Class 2 (9) (10) (11) (12)
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
Class 2: firms with debt rating
us
FV=None Class 1 (1) (2)
M an
Financial Variables
cr
Table 6 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by bond ratings)
0.19 0.09 0.16 [12] [5.4] [5.6] 0.27 [4.1] 0.03 0.05 0.06 0.05 [8.6] [9.2] [11.3] [9.7] -0.03 [-3.1] 0.21 0.28 0.27 0.27
Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.24
0.23
0.07 [3]
0.05 0.01 [3] [0.5] 0.12 [1.1] 0.21 0.23
0.06 [2.9]
0.05 0.02 [3] [1.2] 0.07 [1.4] 0.21 0.22
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
53 Page 53 of 60
ip t
Class 1: firms without debt rating
zit2
0.20 0.20 0.23 0.18 [23.7] [22.9] [20.3] [13.7] 0.07 0.08 [3.9] [3]
FV zit*FV Adj. R2 0.24 0.24 0.22 0.29 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.02 [2.36] Class 1 zit2 -Class 2 zit2 -0.01 [-1.03] Class 1 FV - Class 2 FV qit qit2
0.25 [9.2]
0.22 0.39 0.10 [6.1] [8.9] [5.8] 0.03 0.03 [1.3] [1.6]
qit*FV
(5)
(8)
0.20 [22.7]
0.20 0.20 [22] [22.2] 0.07 [4] -0.01 -0.01 0.34 -0.02 [-0.9] [-0.8] [15.7] [-1.5] 0.02 [1.4] 0.25 0.25 0.10 0.25 Column (6) & Column (10) 0.06 [2.44] -0.01 [-0.99] -0.32 [-8.67] 0.20 [7.7]
0.18 0.16 [5] [5.9] 0.03 [1.1] 0.48 0.48 0.34 0.28 [17.4] [17.4] [15.7] [5.9] 0.21 [5.2] 0.13 0.26 0.10 0.13 Column (6) & Column (10) 0.09 [1.14] 0.02 [2.31] 0.05 [1.54]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.23 0.05 0.05 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.13 [2.87] Class 1 qit2 -Class 2 qit2 0.00 [1.73] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.22 0.14 0.08 [19.7] [9.7] [3.7] 0.07 [2.8] 0.00 0.31 0.52 0.27 [0.1] [6.8] [12.8] [5.7] 0.29 [1.4] 0.23 0.31 0.28 0.32
0.19 0.19 0.19 0.21 0.15 0.16 0.17 0.17 0.19 0.19 0.19 [17] [16.7] [15.1] [13.7] [9.2] [7.2] [18.8] [18.3] [18] [15.4] [13.1] 0.07 0.08 0.06 0.09 [3.8] [2.9] [3.6] [3.1] 0.01 0.01 0.16 0.01 0.01 0.02 0.06 0.02 0.04 0.04 0.19 0.05 0.08 -0.03 [2.2] [1.9] [38.7] [1] [1.9] [3] [11.3] [2.1] [8.2] [8] [23.7] [7.9] [8.9] [-1.5] 0.00 0.00 0.02 [0.2] [0.1] [3] 0.24 0.24 0.31 0.24 0.22 0.30 0.27 0.29 0.25 0.26 0.16 0.25 0.25 0.29 Column (14) & Column (18) Column (22) & Column (26) 0.03 [2.32] -0.02 [-0.57] -0.01 [-1.74] -0.03 [-1.59] -0.01 [-0.24] 0.06 [1.7]
0.33 0.09 0.08 [8.1] [5.8] [3.8] 0.01 [0.4] 0.76 0.43 0.52 0.41 [18.6] [12.2] [12.8] [9] 0.05 [0.6] 0.17 0.17 0.28 0.33
0.12 0.09 -0.08 0.18 0.09 0.06 0.16 0.13 0.15 0.24 [5.3] [2.9] [-2.9] [5.2] [5.5] [2.9] [6] [3.7] [5.4] [6.4] 0.04 0.02 0.03 [1.7] [1.3] [1.3] 0.18 0.18 0.16 0.10 0.23 0.05 0.06 0.04 0.21 0.21 0.19 0.21 0.56 [38.7] [38.7] [38.7] [13.7] [40.1] [10.4] [11.3] [8.7] [22.4] [22.4] [23.7] [11.3] [33.6] 0.07 0.01 0.01 [12.2] [1.2] [0.5] 0.34 0.28 0.31 0.37 0.43 0.43 0.27 0.32 0.17 0.25 0.16 0.17 0.36 Column (14) & Column (18) Column (22) & Column (26) 0.00 [1.5] 0.03 [0.23] 0.01 [1.84] 0.00 [2.08] 0.13 [4.68] 0.13 [0.41]
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: firms with debt rating
M an
Financial Variables
cr
Table 6 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by bond ratings)
0.10 [5.7] 0.03 [1.3] 0.08 [5.5]
0.36
0.16 [9]
0.05 -0.04 [3] [-2] 0.06 [1.5] 0.21 0.29
0.08 [4]
0.05 0.04 [3] [1.5] 0.08 [0.8] 0.21 0.28
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
54 Page 54 of 60
ip t
Class 1: high debt ratio
ait2
0.25 0.25 0.19 0.19 [16.1] [16.2] [15.1] [15.1] 0.26 0.15 [4.8] [3.3]
FV ait*FV Adj. R2 0.19 0.19 0.28 0.28 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.06 [5.07] Class 1 ait2 - Class 2 ait2 0.11 [2.01] Class 1 FV - Class 2 FV rait rait2
0.13 0.14 0.11 0.11 [10.8] [11.6] [11.7] [12.3] 0.15 0.08 [5.7] [4]
rait*FV
(5)
(8)
Adj. R2 0.15 0.16 0.25 0.26 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.02 [2.52] Class 1 rait2 -Class 2 rait2 0.07 [2.7] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.23 0.23 0.13 [15.5] [15.6] [9.7] 0.23 [4.3] 0.06 0.06 0.14 0.12 [4.2] [4.2] [12.5] [10.5] 0.12 [5.6] 0.20 0.20 0.25 0.31 Column (6) & Column (10) 0.06 [2.79] 0.08 [4.51] -0.05 [-7.75]
0.17 0.17 0.24 [13.7] [13.7] [15.1] 0.15 [3.2] 0.11 0.11 0.51 0.06 [9.7] [9.7] [15.5] [4.3] -0.03 [-0.9] 0.30 0.31 0.11 0.20
0.14 0.15 0.08 [11.2] [11.8] [8.1] 0.13 [5] 0.00 0.00 0.14 0.12 [0.2] [0.2] [12.5] [7.5] 0.05 [2] 0.17 0.17 0.25 0.27 Column (6) & Column (10) 0.06 [1.71] 0.07 [0.7] -0.09 [-2.92]
0.08 0.08 0.13 [7.9] [8.4] [10] 0.07 [3.1] 0.10 0.10 0.51 0.01 [8.1] [7.7] [15.5] [0.5] 0.05 [1.3] 0.27 0.27 0.11 0.17
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: low debt ratio
Class 1 (13) (14) (15) (16) 0.19 0.20 [12] [12.1] 0.25 [4.6] 0.03 0.03 [8.6] [8.6]
0.06 [3.7]
0.07 0.08 [5.3] [6.1] 0.12 [4.4] 0.04 0.04 [10.7] [10.1]
0.14 [10.2]
M an
Financial Variables
cr
Table 7 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the debt to capital ratio)
0.04 0.03 [16] [13.3] 0.03 [7.3] 0.21 0.21 0.28 0.33 Column (14) & Column (18) 0.05 [2.77] 0.11 [4.12] 0.00 [4.59]
0.04 0.05 [16] [13.6] 0.04 [8] 0.19 0.19 0.28 0.30 Column (14) & Column (18) 0.01 [1.99] 0.06 [1.54] 0.01 [4.57]
FV=WorkingK_K
Class 2 (17) (18) (19) (20)
Class 1 (21) (22) (23) (24)
0.15 0.15 0.09 [11.8] [11.8] [3.7] 0.13 [3] 0.03 0.03 0.22 0.03 [11.5] [11.4] [40.8] [7.1] 0.04 [1.2] 0.31 0.31 0.40 0.22
0.20 0.20 0.10 0.17 0.17 0.14 [12.4] [12.5] [6.8] [14] [14] [6.7] 0.25 0.13 [4.7] [3] 0.10 0.10 0.05 0.04 0.04 0.04 0.49 0.09 [11] [11] [14.7] [13.3] [12.3] [12.2] [32.7] [9.9] 0.06 0.08 [9.6] [1.6] 0.22 0.23 0.27 0.34 0.32 0.32 0.31 0.23 Column (22) & Column (26) 0.03 [2.45] 0.12 [4.07] 0.06 [1.63]
0.06 0.07 0.07 [6.5] [6.9] [3.8] 0.06 [2.7] 0.03 0.03 0.22 0.04 [11.4] [11] [40.8] [9.8] 0.00 [-0.4] 0.29 0.29 0.40 0.19
Class 2 (25) (26) (27) (28)
0.09 [7.5]
0.10 0.11 0.08 0.09 0.11 [8.3] [9.6] [9.1] [9.5] [6.6] 0.12 0.05 [4.5] [2.6] 0.13 0.12 0.05 0.05 0.04 0.04 0.49 0.13 [13.5] [13] [14.7] [11.8] [11.7] [11.3] [32.7] [13.6] 0.02 -0.03 [3.6] [-1.8] 0.21 0.21 0.27 0.29 0.29 0.29 0.31 0.21 Column (22) & Column (26) 0.01 [1.73] 0.07 [1.21] 0.08 [2.53]
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
55 Page 55 of 60
ip t
Class 1: high debt ratio
zit2
0.23 0.22 [20.3] [19.5] 0.05 [2.5]
0.17 0.16 [19] [18.6] 0.09 [4.9]
FV zit*FV Adj. R2 0.22 0.23 0.30 0.30 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.06 [2.43] Class 1 zit2 -Class 2 zit2 -0.04 [-1.78] Class 1 FV - Class 2 FV qit qit2
0.39 [8.9]
0.32 0.10 0.09 [5.4] [9.6] [6.7] 0.08 0.02 [1.7] [1.8]
qit*FV
(5)
(8)
0.22 0.22 0.16 [19.7] [18.9] [15.8] 0.05 [2.5] 0.00 0.00 0.14 -0.03 [0.1] [0.2] [12.5] [-1.4] 0.06 [2.5] 0.23 0.23 0.25 0.30 Column (6) & Column (10) 0.06 [1.71] -0.04 [-2.4] -0.01 [-3.81] 0.33 [8.1]
0.28 0.10 [5] [9.5] 0.06 [1.5] 0.76 0.76 0.14 0.23 [18.6] [18.5] [12.5] [9.5] 0.07 [4.6] 0.17 0.17 0.25 0.26 Column (6) & Column (10) 0.20 [3] 0.05 [3.13] 0.62 [10.93]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.05 0.23 0.23 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.23 [0.87] Class 1 qit2 -Class 2 qit2 0.06 [1.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.16 0.16 0.22 [16.8] [16.5] [19.2] 0.09 [4.9] 0.01 0.01 0.51 0.00 [0.9] [0.7] [15.5] [-0.2] 0.02 [0.8] 0.30 0.30 0.11 0.23
0.09 [8.4]
0.08 0.11 [6] [2.6] 0.01 [1.5] 0.14 0.14 0.51 -0.36 [10.2] [10.2] [15.5] [-5.2] 1.50 [1.1] 0.26 0.26 0.11 0.28
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: low debt ratio
FV=WorkingK_K
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
Class 1 (21) (22) (23) (24)
0.21 0.21 [13.7] [13.5] 0.05 [2.3] 0.01 0.01 [1.9] [1.7]
0.15 [14.2]
0.19 0.18 0.14 0.15 0.15 0.20 [15.4] [14.9] [13.4] [15.6] [15.3] [13.5] 0.04 0.08 [1.9] [4.6] 0.08 0.08 0.05 0.01 0.02 0.02 0.49 0.09 [8.9] [8.7] [14.7] [2] [5.2] [4.9] [32.7] [8.8] 0.01 -0.03 [1.8] [-1.9] 0.25 0.25 0.27 0.30 0.30 0.31 0.31 0.25 Column (22) & Column (26) 0.04 [2.69] -0.04 [-1.21] 0.06 [10.4]
M an
Financial Variables
cr
Table 7 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the debt to capital ratio)
0.15 [12.8]
0.01 [1] 0.00 [0.2] 0.22 0.23 0.28 0.30 Column (14) & Column (18) 0.06 [2.04] -0.04 [-2] 0.00 [6.08] 0.18 0.10 [5.2] [2.1] 0.09 [2.6] 0.23 0.23 [40.1] [40.1]
0.04 [16]
0.13 [10.5]
0.04 0.06 [16] [12.6] 0.02 [6.1] 0.43 0.43 0.28 0.29 Column (14) & Column (18) 0.02 [1.88] 0.08 [2.13] 0.20 [13.62]
0.15 0.21 [14] [11.9] 0.09 [4.8] 0.01 0.01 0.22 0.01 [2.6] [2.4] [40.8] [1.5] 0.00 [-0.4] 0.30 0.30 0.40 0.22
0.09 0.08 -0.68 [8.4] [5.9] [-18.9] 0.01 [1.6] 0.04 0.04 0.22 0.01 [13] [12.9] [40.8] [1.9] 0.28 [0.9] 0.28 0.28 0.40 0.64
0.24 [6.4]
0.16 [3.2] 0.08 [2.2] 0.56 0.56 0.05 [33.6] [33.7] [14.7]
Class 2 (25) (26) (27) (28)
0.09 [8.3]
0.09 [8.2]
0.05 [6] 0.01 [1.3] 0.36 0.36 0.27 0.25 Column (22) & Column (26) 0.09 [2.21] 0.07 [3.25] 0.52 [6.68]
0.04 [9.4]
0.25
0.07 -0.37 [5.8] [-10.5] 0.01 [1.5] 0.04 0.49 -0.08 [9.4] [32.7] [-3.5] 0.83 [0.5] 0.25 0.31 0.57
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
56 Page 56 of 60
ip t
Class 1: Equity-dependent firms
ait2
0.24 0.24 0.12 0.12 [16.4] [16.5] [14.9] [14.9] 0.14 0.19 [2.8] [7.1]
FV ait*FV Adj. R2 0.19 0.19 0.25 0.27 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.12 [2.25] Class 1 ait2 - Class 2 ait2 -0.05 [-1.71] Class 1 FV - Class 2 FV rait rait2
0.13 0.14 0.11 0.11 [11.8] [12.9] [9.6] [9.8] 0.14 0.06 [6.1] [2.3]
rait*FV
(5)
(8)
0.24 [15.9]
Adj. R2 0.17 0.18 0.22 0.22 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.03 [2.91] Class 1 rait2 -Class 2 rait2 0.08 [1.77] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.24 0.24 [16] [15.5] 0.11 [2.4] 0.06 0.06 0.49 0.06 [4.4] [4.4] [15.4] [4.3] 0.01 [2.3] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.13 [2.33] -0.07 [-2.78] 0.04 [4.9]
0.11 0.12 0.11 [14.5] [14.6] [13.7] 0.19 [7.1] 0.02 0.02 0.04 0.02 [1.9] [1.7] [3.6] [2] -0.03 [-1.2] 0.25 0.27 0.16 0.25
0.14 0.15 0.13 [12.7] [13.6] [11.5] 0.12 [5.5] -0.01 -0.01 0.49 -0.01 [-0.6] [-0.7] [15.4] [-0.3] 0.06 [2.2] 0.19 0.19 0.10 0.19 Column (6) & Column (10) 0.08 [2.58] 0.08 [0.51] -0.11 [-0.31]
0.07 0.07 0.08 [5.8] [5.9] [5.9] 0.04 [1.5] 0.10 0.10 0.04 0.12 [6.8] [6.6] [3.6] [6.5] -0.04 [-1.4] 0.23 0.23 0.16 0.23
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: Not equity-dependent firms
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.21 0.22 0.18 0.14 0.12 0.04 0.23 0.24 0.18 0.11 0.11 0.11 [12.7] [12.8] [3.5] [10.7] [10.8] [1.8] [12.8] [12.8] [8.1] [13.5] [13.5] [6.5] 0.14 0.20 0.14 0.22 [2.8] [6.5] [2.9] [6.8] 0.03 0.03 0.23 0.02 0.04 0.03 0.04 0.04 0.11 0.11 0.51 0.10 0.04 0.04 0.05 0.05 [8] [7.9] [41.4] [6.4] [12.9] [12.5] [16.4] [14.7] [12.3] [12.4] [33.9] [11.6] [11.7] [11.5] [13.4] [12.5] 0.04 0.04 0.04 0.07 [6] [0.6] [1.9] [0.8] 0.21 0.21 0.41 0.22 0.30 0.31 0.24 0.32 0.23 0.24 0.32 0.24 0.29 0.30 0.21 0.31 Column (14) & Column (18) Column (22) & Column (26) 0.10 [3.35] 0.13 [3.23] -0.06 [-2.84] -0.08 [-3.31] 0.00 [3.52] 0.06 [10.99]
M an
Financial Variables
cr
Table 8 -Response of investment to alternative fundamentals and financial variables (when firms are grouped by their KZ index)
0.08 0.10 0.07 0.03 0.03 0.14 0.09 0.10 0.11 [7.3] [8.3] [3.9] [2.3] [2.4] [8.9] [8.6] [9.4] [7.7] 0.11 0.03 0.10 [4.8] [1] [4.3] 0.04 0.03 0.23 0.04 0.04 0.04 0.04 0.07 0.13 0.13 0.51 0.14 [9.4] [8.6] [41.4] [8.1] [15.2] [15.1] [16.4] [18.5] [14.7] [14] [33.9] [14.8] 0.01 -0.05 0.03 [0.8] [-10.5] [2.1] 0.20 0.20 0.41 0.20 0.28 0.28 0.24 0.31 0.23 0.24 0.32 0.24 Column (14) & Column (18) Column (22) & Column (26) 0.07 [2.53] 0.02 [1.82] 0.09 [1.97] 0.06 [1.74] -0.01 [-3.43] 0.09 [11.47]
0.08 0.08 0.10 [7] [7.1] [7.6] 0.04 [1.3] 0.04 0.04 0.05 0.05 [9.9] [9.7] [13.4] [10.1] -0.02 [-3.2] 0.25 0.25 0.21 0.25
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the ones taking place in the higher (lower) 50 percentile of the firms when they are sorted by the size of their KZ index.
57 Page 57 of 60
ip t
Class 1: Equity-dependent firms
zit2
0.21 0.21 0.19 0.18 [19.7] [19.2] [19.7] [19.1] 0.03 0.13 [1.4] [6.5]
FV zit*FV Adj. R2 0.22 0.22 0.29 0.30 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.02 [2.26] Class 1 zit2 -Class 2 zit2 -0.10 [-2.66] Class 1 FV - Class 2 FV qit qit2
0.35 [8.7]
0.27 0.11 0.09 [5.5] [9.5] [5.7] 0.11 0.02 [2.6] [2.5]
qit*FV
(5)
(8)
Adj. R2 0.05 0.05 0.21 0.22 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.19 [1.87] Class 1 qit2 -Class 2 qit2 0.08 [1.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.20 0.20 0.20 [19.2] [18.6] [19.1] 0.03 [1.4] 0.00 0.00 0.49 0.00 [0.2] [0.3] [15.4] [0.3] 0.08 [3.3] 0.23 0.23 0.10 0.22 Column (6) & Column (10) 0.02 [2.61] -0.10 [-3.69] 0.01 [2.03]
0.19 0.18 0.18 [17.5] [17.2] [19.1] 0.13 [6.5] 0.00 0.00 0.04 -0.05 [0.2] [-0.1] [3.6] [-2.4] 0.00 [0.3] 0.29 0.30 0.16 0.29
0.28 0.22 0.09 [7.4] [4.8] [2.5] 0.07 [1.9] 0.84 0.84 0.49 -0.24 [19.8] [19.7] [15.4] [-3.6] 1.53 [20.4] 0.18 0.18 0.10 0.30 Column (6) & Column (10) 0.15 [1.89] 0.05 [2.53] 0.72 [19.62]
0.10 0.08 0.11 [8.6] [5.1] [8.8] 0.02 [2.4] 0.11 0.11 0.04 0.16 [8.6] [8.6] [3.6] [6.5] -0.04 [-2.3] 0.24 0.24 0.16 0.24
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: Not equity-dependent firms
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.20 0.19 0.21 0.17 0.16 0.16 0.21 0.22 0.19 0.18 0.17 0.16 [13.6] [13.5] [19.1] [13.9] [13.8] [19.1] [14.5] [14.3] [19.1] [16.8] [16.4] [19.1] 0.03 0.12 0.02 0.12 [1.3] [6.3] [0.8] [6.3] 0.01 0.01 0.23 0.02 0.01 0.01 0.04 0.00 0.09 0.09 0.51 0.10 0.02 0.01 0.05 0.00 [1.5] [1.4] [41.4] [2] [3.2] [2.8] [16.4] 0.00 [10.3] [10.2] [33.9] [10.5] [3.6] [3.2] [13.4] [0.6] 0.01 0.01 0.03 0.02 [2.4] [1.3] [2.7] [1.3] 0.22 0.22 0.41 0.22 0.29 0.30 0.24 0.29 0.25 0.25 0.32 0.25 0.29 0.30 0.21 0.29 Column (14) & Column (18) Column (22) & Column (26) 0.03 [2.03] 0.05 [1.76] -0.10 [-4.34] -0.11 [-4.5] 0.00 [4.97] 0.07 [9.79]
M an
Financial Variables
cr
Table 8 (cont'd) -Response of investment to alternative fundamentals and financial variables (when firms are grouped by their KZ index)
0.14 0.08 -0.58 [4.7] [2.1] [-18.9] 0.09 [2.8] 0.25 0.25 0.23 0.02 [42.8] [42.8] [41.4] [2.9] 0.29 [38.3] 0.46 0.46 0.41 0.66 Column (14) & Column (18) 0.01 [2.75] 0.07 [2.53] 0.22 [18.85]
0.09 0.07 0.12 [8.3] [4.9] [8.7] 0.02 [2.2] 0.04 0.04 0.04 0.05 [14] [13.9] [16.4] [11.2] -0.01 [-3.4] 0.27 0.27 0.24 0.27
0.21 [6.1]
0.13 -0.32 [3.2] [-10.7] 0.10 [2.8] 0.57 0.56 0.51 -0.08 [34] [34.1] [33.9] [-3.7] 0.87 [37.5] 0.36 0.36 0.32 0.60 Column (22) & Column (26) 0.06 [4.61] 0.08 [3.58] 0.53 [28.9]
0.09 0.07 0.10 [7.9] [4.6] [7.7] 0.02 [2.3] 0.04 0.04 0.05 0.05 [9.5] [9.5] [13.4] [5.4] -0.01 [-1.2] 0.24 0.24 0.21 0.24
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the ones taking place in the higher (lower) 50 percentile of the firms when they are sorted by the size of their KZ index.
58 Page 58 of 60
ip t
ait
Size of capital Firms with small capital stock (1) stock Firms with large capital stock (2)
rait
zit
-2 stdev -1 stdev +1 stdev +2 stdev -0.545 -0.270 0.281 0.557
-2 stdev -1 stdev +1 stdev +2 stdev -0.570 -0.277 0.309 0.602
us
-2 stdev -1 stdev +1 stdev +2 stdev -0.410 -0.205 0.204 0.408
Shock values
cr
Table 9 - Joint effects of linear and nonlinear terms of fundamentals in investment equations
-0.042
0.056
0.126
-0.044
-0.028
0.043
0.097
-0.088
-0.050
0.070
0.151
(1)-(2)
-0.015 -0.056
-0.019 -0.023
0.041 0.015
0.106 0.021
-0.016 -0.028
-0.018 -0.010
0.038 0.004
0.096 0.001
-0.083 -0.006
-0.045 -0.005
0.060 0.011
0.126 0.026
Number of employees
Firms with low number of employees (3) Firms with high number of employees (4) (3)-(4)
-0.073 -0.008 -0.066
-0.043 -0.017 -0.026
0.055 0.044 0.011
0.122 0.114 0.008
-0.043 -0.011 -0.033
-0.028 -0.015 -0.013
0.043 0.037 0.006
0.098 0.094 0.004
-0.092 -0.073 -0.019
-0.050 -0.041 -0.009
0.069 0.059 0.010
0.146 0.127 0.020
Dividend payout ratio
Firms with low dividend payout ratio (5) Firms with high dividend payout ratio (6) (5)-(6)
-0.068 -0.021
-0.040 -0.024
0.052 0.050
0.116 0.126
-0.042 -0.027
-0.028 -0.022
0.043 0.041
0.099 0.099
-0.104 -0.048
-0.055 -0.034
0.069 0.063
0.143 0.146
-0.048
-0.017
0.002
-0.010
-0.015
-0.006
0.002
0.000
-0.056
-0.020
0.006
-0.003
Firms with low dividend to capital ratio (7) Firms with high dividend to capital ratio (8) (7)-(8)
-0.069 -0.015 -0.053
-0.039 -0.024 -0.015
0.048 0.056 -0.008
0.105 0.144 -0.039
-0.040 -0.024 -0.017
-0.027 -0.020 -0.008
0.044 0.037 0.007
0.102 0.089 0.013
-0.099 -0.062 -0.037
-0.053 -0.037 -0.016
0.070 0.056 0.014
0.147 0.124 0.023
Firms without bond rating (9) Firms with bond rating (10) (9)-(10)
-0.065 0.002 -0.067
-0.040 -0.009 -0.031
0.055 0.029 0.026
0.125 0.078 0.047
-0.042 0.024 -0.066
-0.028 0.001 -0.028
0.043 0.023 0.020
0.099 0.068 0.032
-0.091 -0.056 -0.035
-0.050 -0.033 -0.017
0.068 0.050 0.017
0.143 0.111 0.032
-0.057 -0.045
-0.038 -0.028
0.057 0.041
0.133 0.093
-0.041 -0.026
-0.030 -0.018
0.051 0.029
0.122 0.067
-0.106 -0.062
-0.056 -0.037
0.072 0.058
0.150 0.129
-0.012
-0.010
0.016
0.039
-0.014
-0.012
0.023
0.055
-0.044
-0.019
0.014
0.021
-0.080
-0.045
0.054
0.118
-0.098
-0.057
0.077
0.170
-0.101
-0.058
0.086
0.187
-0.016
-0.016
0.032
0.079
-0.007
-0.018
0.048
0.123
-0.005
-0.018
0.054
0.139
-0.064
-0.029
0.023
0.039
-0.091
-0.039
0.030
0.047
-0.096
-0.040
0.032
0.049
Mean
Standard deviation
Min
Max
Bond rating
Firms with high debt to capital ratio (11) Firms with low debt to capital ratio (12) (11)-(12)
KZ index
Equity-dependent firms (13) Not equity-dependent firms (14) (13)-(14)
Ac
ce pt
Total debt to capital ratio
ed
Dividend to capital ratio
M an
-0.071
Fundamentals ait rait zit
0.00 0.01 0.02
0.20 0.28 0.29
-0.69 -1.13 -1.13
0.70 0.94 1.14
25th 50th 75th percentile percentile percentile -0.12 -0.13 -0.14
0.00 0.01 0.02
0.11 0.15 0.16
Note: The gray-shaded rows are for the firms expected to be financially constrained. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), and zit is the mandated investment rate. stdev stands for standard deviation. Shock values are calculated as: (mean of fundamental +standard deviation of fundamental* degree of standard deviation) where the degree can be -2, -1, 1 or 2. The standard deviationas and means of each fundamental is given at the bottom of the table. The joint effects of linear and nonlinear terms of fundamentals are calculated as: b 1*shock value+b 2*(shock value)2 where b 1 and b 2 are the estimated coefficients from iit b1xit b2 xit2 cCF_ Kit T F uit , taken from columns (6) and (10) of Tables 2-8. In these calculations, the empirical specification includes cash flow to capital ratio as financial variable.
59 Page 59 of 60
Research Highlights
d
M
an
us
cr
ip t
Investment/fundamental sensitivity of firms with/without financial problems New fundamentals are used: profitability shocks and mandated investment rate Investment is regressed on fundamentals and financial variables for two groups Financially constrained firms are found to be more responsive to fundamentals Results support expectations of contracting models of financial market imperfections
Ac ce pt e
• • • • •
Page 60 of 60
S0148-6195(14)00031-9 http://dx.doi.org/doi:10.1016/j.jeconbus.2014.05.001 JEB 5683
To appear in:
Journal of Economics and Business
Received date: Revised date: Accepted date:
10-10-2013 28-5-2014 30-5-2014
Please cite this article as: Bayraktar, N.,Fixed Investment/Fundamental Sensitivities under Financial Constraints, Journal of Economics and Business (2014), http://dx.doi.org/10.1016/j.jeconbus.2014.05.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ip t
us
cr
Fixed Investment/Fundamental Sensitivities under Financial Constraints
Nihal Bayraktar Pennsylvania State University - Harrisburg
te
d
M
an
While most models with financial market imperfections predict investment by financially constrained firms to be more sensitive to financial variables, contracting models argue that investment by such firms should be more sensitive to fundamental determinants of investment because fundamentals capture both investment opportunities and changes in the financial position. By first grouping U.S. manufacturing firms as either financially constrained or unconstrained, this paper examines systematic differences in investment/fundamental sensitivities. The findings show that, as expected of contracting models, investment by financially constrained firms is more responsive to fundamentals. These fundamentals are captured by two prominent empirical measures: profitability shocks and mandated investment rate.
Ac ce p
Keywords: financial market imperfections, contracting models, fixed investment, fundamentals, profitability shocks JEL Classification Number: D2, E22, G1, G3
1 Page 1 of 60
1
Introduction
The investment literature identifies fundamental determinants of investment as one of the main
ip t
determinants of fixed capital investment at the firm level. However, there are different views on investment/fundamental sensitivities of, especially, financially constrained firms. For example,
cr
many models in the financial market imperfections literature expect investment by financially
us
constrained firms to be less sensitive to fundamentals but more responsive to financial variables. Conversely, according to contracting models, which study costly borrowing of firms in the
an
presence of financial frictions, financially constrained firms’ investment behavior is predicted to be more responsive to fundamentals when compared to the investment behavior of financially
M
unconstrained firms. This is due to the fact that changes in fundamentals capture changes in not only investment opportunities but also financial positions. If the shock to fundamentals is
d
persistent, it improves the expected marginal benefit of investment. However, for the same
te
reason, it also improves the terms of trade in financial transactions of the firm. If the expected
Ac ce p
profit increases, the probability of default declines with the increasing value of the firm. Therefore, the agency cost, which compensates risk neutral lenders for expected loss from default, must also decrease.1
This paper, in light of such arguments in the literature, tries to understand whether or not
1
This negative relationship between the profitability shock (one measure of fundamentals) and the agency cost is
firmly established by the literature on defaultable debt, for instance, Carlstrom and Fuerst (1997), Bernanke et al. (2000), Cooley and Quadrini (2001), and Hennessey and Whited (2007) for corporate finance, Chatterjee et al. (2007) for consumer finance, Marcet and Marimon (1992) and Cooley et al. (2004) for long term contract. In the literature, it is clear that default history and current earnings are the most important factors in determining credit limits and interest rates for any unsecured debt financing in reality.
2 Page 2 of 60
investment by expected-to-be financially constrained firms really responds more strongly to shocks to fundamentals than investment by financially unconstrained firms. When compared to previous empirical studies, one main contribution of the paper is the use of alternative measures
ip t
of fundamentals. In the literature, there are many concerns about the predictive power of one of the most commonly used fundamental determinants of fixed capital investment: Tobin’s q (the
cr
ratio of asset market value of a firm to its replacement cost of capital). Alternatively, empirical
us
papers based on investment models with non-convex adjustment costs introduce new measures of fundamentals. These present a forward-looking behavior – an important feature that can allow
an
fundamentals to better capture investment opportunities. In this paper two of these are used as fundamental determinants of investment: profitability shocks and the mandated investment rate
d
included for the purpose of comparison.
M
(a gap measure between the desired and actual capital stocks).2 In the analyses, Tobin's q is also
te
The analyses in the paper are based on a reduced form investment equation, in which both fundamental determinants of investment and financial variables are taken as explanatory
Ac ce p
variables, as well as their interaction terms in some specifications. The regressions are separately run for financially constrained and unconstrained firms. Based on the regression results, systematic differences in the estimated coefficients of the fundamental variables of financially constrained and unconstrained firms are investigated. A firm-level panel data set is constructed from the COMPUSTAT database. The set includes U.S. manufacturing firms for the period of 1983-1996. Different firm characteristics are used to identify financially constrained firms. The
2
“Fundamental Q” would be another good candidate (Gilchrist and Himmelberg, 1995 and 1998; Del Boca et al.,
2008). It has not been included in the study because it has been already reported in the literature that the significance of financial variables drops and fundamentals get more significant with "Fundamental Q."
3 Page 3 of 60
criteria are the level of capital stock, number of employees, dividend to capital ratio, dividend payout, debt-to-capital ratio, firms' bond rating, and the KZ (Kaplan and Zingales) index. The ratio of cash flow to capital, sales to capital, and working capital to capital are the financial
ip t
variables included in the analyses.
The empirical findings support the prediction of contracting models. Firms with financial
cr
constraints exhibit a stronger investment-fundamental sensitivity when compared to the group of
us
firms that are less constrained financially. Even though it is not the main purpose of the paper, in the regression outcomes we can also observe investment/financial variable sensitivities across
an
firm classifications because the regression specifications already include financial variables in addition to fundamentals. In the analysis, it can be seen that the investment/financial variable
M
sensitivity of financially constrained firms is lower in many cases than the investment/financial
d
variable sensitivity observed in the group of unconstrained firms.
te
The rest of the paper is organized as follows. Section 2 presents a contracting model of investment. Section 3 gives information about the link between investment, fundamentals, and
Ac ce p
financial variables, and highlights some empirical issues. Section 4 presents some details on the data set and variables, including the construction of the fundamentals. In Section 5, the empirical results are presented. Section 6 concludes.
2 A Contracting Model of Investment One of the main models of investment, the Q theory, presents a formal link between a firm's investment and marginal q. Furthermore, it shows that marginal q should be the sole determinant of investment. However, the explanatory power of Tobin's q (empirical marginal q) in the investment process has been found to be negligible in many empirical studies; and thus the literature started searching for new investment models. 4 Page 4 of 60
The introduction of financial market imperfections in investment models is one such examples. The financial market imperfections literature assumes that firms' net worth determines their financial position. On the one hand, a firm with low net worth can be considered financially
ip t
constrained, since it is likely to face an asymmetric information problem in financial markets, which makes it difficult for them to find enough external funds to finance their investment
cr
projects. Because of this, their investment decisions are expected to be highly correlated with
us
their internal funds, but not fundamentals. Firms with high net worth, on the other hand, are expected to have a smaller asymmetric information problem; thus, they can find enough external
an
funds to finance their capital adjustment and follow the investment process suggested by changes in fundamentals.
M
Contracting models, a type of financial market imperfections model, also study costly
d
borrowing of firms in the presence of financial frictions. Improving fundamentals, such as
te
profitability of firms, signal not only improvements in investment opportunities but also improvements in firms’ financial position through easier borrowing conditions and declining
Ac ce p
agency costs. This in turn leads to an even higher level of investment. In light of this argument, investment of financially constrained firms is expected to be more sensitive to fundamentals. The following model introduces financial frictions through contracts. The aim is to determine why improving fundamentals may lead to healthier financial position of firms. The model combines the costly state verification framework by Townsend (1978) and Cooley and Quadrini (2001) with the firm-level investment model introduced by Bayraktar (2009). The model creates financial frictions in the market through the costly state verification. The firm honors debt obligations only when the value of taking this action is greater than the value of defaulting. In the
5 Page 5 of 60
occasion of default, the investor verifies the firm's revenue, which is costly but eliminates the incentive problem. The lender, who lends R.Bt+1 today, gets Bt+1 tomorrow if the firm does not default, where R
ip t
is the discounted price of the bond less than one. It is taken as a function of expected profit shocks, capital, and debt. If the firm defaults, the lender renegotiates the debt burden of the firm
cr
in a way to make the firm indifferent with regard to continuing the project or exiting the market.
an
us
The value function, V(⋅), is defined as:
subject to
. In the function, t denotes time period,
(1) (⋅) is the present
M
discounted future value of the firm, C(⋅) is the investment cost function, It stands for investment, Kt is the current capital stock, δ is the depreciation rate,
(⋅) is the profit function, β is the fixed
d
discount factor, and At is the profitability shock. Since profitability shocks are highly auto-
(⋅) are homogenous of degree one in investment and capital, and
Ac ce p
assumed that both C(⋅) and
te
correlated, the expected value of future profitability shock (EAt+1) given At can be calculated. It is
C(⋅) is a convex function. Let
denote the renegotiated debt, which becomes effective if the firm defaults. The
renegotiated debt burden must satisfy the condition of the firm’s indifference between continuing the project and exiting the market: 0 = V(EAt+1|At, Kt+1,
). The reservation utility is assumed
to be zero without loss of generality. The condition implicitly defines the renegotiated debt as a function of At+1 and Kt+1. Due to the possibility of default, the discounted price of the bond, R(⋅), must satisfy the following zero profit condition to convince a risk neutral lender to hold the risky debt:
6 Page 6 of 60
(2) where prob is the probability of no-default corresponding to the cases where the value of At+1 is > 0 (i.e. no-default set).
is the verification cost
ip t
high enough to produce
cost. Given
and >0, in this case,
.
is an increasing function of expected
us
It is straightforward to show that
cr
per unit of debt issue. In case of default, the lender receives renegotiated debt minus verification
profitability shocks given At (i.e.
, which is the main fundamental measure of
an
investment in the empirical part of this paper. In other words, the agency cost, because, for any given set of
M
⁻¹ ≤ β⁻¹, is a decreasing function of
.
d
choices made by the firm (Kt+1, Bt+1), the probability of default is decreasing in
te
3 Investment and Fundamentals: Empirical Issues
Ac ce p
The contracting model presented in the previous section indicates that the profitability shock is a shock to both fundamentals and to finance. Essentially, this means that the shock to the profit cannot be classified solely as a fundamental factor, or other way of saying, as a factor that does affect only the profit but not the finance. This is important because this possibility has not been considered in many empirical studies in the literature. For example, one of the common structures used in the cash-flow/investment sensitivity literature is a regression equation given by
inv = bx + cFV + u,
(3)
where inv is the investment rate. While x is a fundamental measure used to capture investment opportunities (mostly Tobin’s q), and FV stands for financial variables such as net worth or 7 Page 7 of 60
internal funds. u is the error term. This regression equation estimates the sensitivity of investment to changes in internal funds, FV, controlling for investment opportunities, x. In the literature, the equation is estimated separately for financially constrained and unconstrained
ip t
firms which are identified based on a priori proxies.
The capital imperfections literature, on the one hand, has shown that the sensitivity of
cr
investment to financial variables is higher for financially constrained firms, and FV is significant
us
but x is not. On the other hand, another group of researchers have found that finance does not matter, since x is significant but FV is not. While the second group criticizes the first group by
an
arguing that the source is mismeasurement of x, the first group criticizes the second group by arguing that the source is mismeasurement of FV. This seems to be an almost endless debate in
M
the investment literature. Even when a researcher observes x without any measurement error, the
d
researcher may not be able to escape this endless debate.
te
Let us consider two different firms to highlight the view point of the contracting models: one firm financially constrained and the other financially unconstrained. For the financially
Ac ce p
unconstrained firm, a good shock today affects the investment decision only through its effect on the expected marginal efficiency of capital tomorrow. However, for the financially constrained firm, the good shock affects the investment decision not only through the effect on the expected marginal efficiency of capital but also through the effect on the agency cost of borrowing today. In this case, financial frictions, if exist, provide an amplification effect through the agency cost channel. In this regard, two predictions can be made. First, if a researcher correctly identifies the shock and regresses investment on the shock variable, x, the coefficient must be significant regardless of whether or not the firm is financially constrained, assuming the shock is persistent.
8 Page 8 of 60
This presents another interpretation issue: is the coefficient significantly positive because a good shock improves the forecast of fundamentals, decreases the agency cost of borrowing or both? Second, if one firm is financially constrained but the other is not, the coefficient of the
ip t
fundamental shock variable might be greater for the financially constrained firm due to the additional amplification channel of the agency cost of contracting models.3
cr
At this point, we need to remember that if the profitability shock is correctly identified, its
us
coefficient in an empirical investment equation must convey information about both the fundamentals and the finance variable. Therefore, one cannot infer that the finance does not
an
matter since x is significant but FV is not (or less significant than x), because the effect of financial frictions must show up in the coefficient for x, among others, if there are financial
M
frictions. Thus, in order to study how financial frictions affect the investment decision, it is
d
necessary to check whether or not there are systematic differences in the coefficient of x among
te
firms with different characteristics instead of only testing the significances of x versus FV. In this regard, it is expected that firms with similar profitability shocks (x) can produce a different level
Ac ce p
of investment due to disparities in their financial positions. In light of this discussion, the paper analyzes systematic differences between financially constrained and unconstrained firms in their investment sensitivity to fundamentals, including profitability shocks. Another source of disagreement in the literature is how to appropriately measure fundamentals. Despite the availability of a large range of financial variables to determine firms' financial position, most empirical studies use only Tobin's q as a proxy for investment 3
The predictions made from a development in financial frictions literature, for example, Cooley and Quadrini
(2001) in the framework of short-term standard debt contract, and Clementi and Hopenhayn (2006), Albuquerque and Hopenhayn (2004) and Cooley et al. (2004) in the framework of long-term contract show that investment of financially constrained firms responds more to fundamentals or profitability shocks.
9 Page 9 of 60
opportunities. Theoretically, the correct measure to capture investment opportunities is marginal q, which is defined as the present discounted value of future profits generated by an additional unit of capital. Since marginal q is not empirically observable, different substitutes such as
ip t
Tobin’s q (ratio of firm's average value to its capital stock) are introduced in empirical studies. These empirical results show that the explanatory power of Tobin's q in investment equations is
cr
much weaker compared to financial variables. A possible reason for this failure would be the
us
inadequacy of Tobin's q in capturing investment opportunities due to measurement errors.4 Alternative measures of fundamentals are introduced in empirical studies based on non-
an
convex capital adjustment cost models.5 Two examples are profitability shocks and the gap measure between the desired and actual capital stock. In this paper, these two fundamentals are
M
used as proxies for investment opportunities. One big advantage of these fundamentals is that,
d
even though they are constructed using current variables, they present a forward-looking
te
behavior. Since profitability shocks are serially correlated, the current value of shocks gives information about future profitability. The gap measure between the desired and actual capital
Ac ce p
stocks is also informative about future investment behavior, because the magnitude of the gap determines whether a firm invests in the current period or in the future. Another advantage of these fundamentals is that since they give information about the future, the correlation between the fundamentals and current internal funds is low. For example, the correlation between the profitability shocks and the cash flow to capital ratio is only 0.25 as shown in Bayraktar (2002a). 4
The following papers are some of the examples focusing on measurement errors associated with fundamentals:
Kaplan and Zingales (1995 and 1997), Gilchrist and Himmelberg (1995 and 1998), Gomes (2001), Erickson and Whited (2000), Cooper and Ejarque (2001), and Abel and Eberly (2003). 5
See Caballero, Engel and Haltiwanger (1995), Cooper and Haltiwanger (2006), Cooper and Ejarque (2001),
Bayraktar (2002b), Bayraktar (2009), and Bayraktar et al. (2005).
10 Page 10 of 60
4 Data and Construction of Fundamentals The main data source is the COMPUSTAT firm-level database. The data set, covering the period
ip t
from 1981 to 1996, includes U.S. manufacturing firms.6 The total number of firms is 463 and the total number of panel observations is 6450, where the balanced panel would have had 6482
us
cr
observations.7 The following sub-sections introduce main variables used in the study.
4.1 Investment, Financial Variables, and Firm Characteristics
an
The definition of capital includes plant, property, and equipment (PPE). Investment is defined as capital expenditure net of capital sales, including capital retirements. The replacement value of
M
capital is calculated using a perpetual inventory method: Kt = (1-δ) Kt-1 + It, where Kt is the real capital stock, It is real investment, which is calculated by deflating the nominal value by the 4-
d
digit investment price index. δ is the 2-digit depreciation rate from the Bureau of Labor Statistics
te
(BLS) database. It is equal to the average value of the depreciation rates for the period of 1981-
Ac ce p
1996. The investment rate is defined as the ratio of real investment to the replacement value of capital. Since investment is net of sales of capital, there are also negative investment rates, corresponding to nearly 10 percent of observations. The following financial variables capture the effects of internal funds on investment: cash flow to capital ratio (ratio of the book value of cash flow to the beginning of period book value of gross total PPE); sales to capital ratio (ratio of the book value of net sales revenue to the beginning of period book value of PPE); working capital to capital ratio (ratio of the book value 6
Since the retirement data, used in constructing investment series, which were not reported since 1996, the
following years are not included in this study. Some other variables used in constructing the fundamentals are not available for more recent years. 7
Detailed information on sample selection is in Appendix A.
11 Page 11 of 60
of working capital –the difference between current assets and liabilities– to the beginning of period book value of PPE. In the literature different firm characteristics are used in determining firms with possible
ip t
financial restrictions. The following a priori proxies are used in identifying financially constrained firms in this paper: book value of PPE stock; number of employees; total debt to
cr
capital ratio (ratio of the book value of total debt to the beginning of period PPE); bond rating
us
(the issuer’s senior debt rating assigned by Standard & Poor's); dividend payout (ratio of the book value of dividends to the book value of income before extraordinary items); dividend to
an
capital ratio (ratio of the book value of dividends to the beginning of period book value of PPE). In addition to these a priori proxies, an index measure of equity dependence of firms (the KZ
M
index), which is initially introduced by Kaplan and Zingales (1997), is applied in classification of
te
4.2 Fundamentals
d
firms.
Ac ce p
Two different measures of fundamentals are introduced in the paper: profitability shocks and the mandated investment rate. Besides these fundamentals, Tobin's q is also included for the purpose of comparison.
4.2.1 Profitability Shocks
The first set of fundamentals capturing investment opportunities is idiosyncratic profitability shocks. Cooper and Haltiwanger (2006) display that the empirical relationship between the investment rate and profitability shocks is nonlinear and asymmetric such that the response of investment to positive shocks is much stronger than its response to negative shocks which can be explained with the presence of both convex and nonconvex adjustment costs. The profitability shocks are presented in the following firm-level profit function: 12 Page 12 of 60
(4) where
is the profitability shock, consisting of both aggregate and idiosyncratic components,
is the curvature of the profit function, and .
ip t
alternative ways to calculate
is the firm level capital stock. There are two
by regressing the log of real profits on the log of real capital using firm-level panel
us
calculates
The first method
cr
Profitability shocks as residuals from a profit function regression
data. Fixed effects are removed and time dummies are included to identify the effects of
an
aggregate profitability shocks.
ln (.) ln Kit F T rait ,
M
(5)
where rait (firm-level error term) is taken as idiosyncratic profitability shocks and named as
d
residual profitability shocks (rait) from now on.
A second, indirect,
is to consider the first order condition for profit maximization with
Ac ce p
method of calculating
te
Profitability shocks as residuals from first order conditions
respect to employment. Since the employment series are more reliable, this allows us to avoid possible measurement errors in profit data. It is assumed that a firm maximizes the following profit function with respect to labor:
( A, K ) max R( Aˆ , K , L) Lw, L
(6)
where A is the profitability shock that contains both aggregate and idiosyncratic shocks, Aˆ is the shock to the revenue function, K is the capital stock, L is labor, and w represents labor wage. It should be noted that these shocks have been calculated for each firm and period. In order to
13 Page 13 of 60
simplify the equations, time and firm subscripts are removed. Assuming that the product market is imperfectly competitive, the revenue function is defined as
R( Aˆ , K , L) Aˆ py Aˆ y y Aˆ y1 ,
ip t
(7)
where the constant returns to scale Cobb-Douglas production function is assumed to be
cr
y AK K L L , where αL and αK are the production function coefficients for labor and capital,
us
respectively. The demand curve is p y , where ξ is the elasticity of the demand curve.
an
From the first order condition with respect to L, the optimum value of L is:
1
L (1 )
K
K (1 ) 1 L (1 )
. (8)
M
Aˆ (1 ) L L* w
After plugging this optimum labor back into the profit equation, it becomes:
te
d
( A, K ) ( A1 A2 ) K ,
(9)
L (1 )
1
Ac ce p
Aˆ (1 ) L 1 L (1 ) Aˆ (1 ) L 1 L (1 ) (1 L )(1 ) , A2 w . where , A1 Aˆ w w 1 L (1 ) Note that ( A1 A2 ) is the profitability shock, A, which is equal to L (1 ) L (1 ) 1 1 1 L (1 ) 1 L (1 ) 1 L (1 ) 1 L (1 ) ˆ A A w [(1 ) L ] [(1 ) L ] .
(10)
From Equation (8) Aˆ is calculated as
w Aˆ (1 ) L
L* K
L (1 )
. (11)
14 Page 14 of 60
In terms of calibration, is estimated by regressing the log of real net profit data on the log of the replacement value of real capital using firm-level panel data after removing fixed effects. Its value is estimated at 0.61. αL is estimated as 0.73 again using firm-level data.8 ξ is obtained (1 L )(1 ) . These coefficients imply a demand elasticity of -0.15 and a markup of 1 L (1 )
ip t
from
cr
about 18 percent.
us
The aggregate shocks are calculated as the annual mean of A across firms, and the firm-level idiosyncratic component of A, presented by ait, is taken as the deviation from this mean. ait is
an
named, from now on, as profitability shocks from the first order condition. The correlation
M
coefficient between rait and ait is calculated as 0.55 in Bayraktar (2002a).
4.2.2 Mandated Investment Rate
d
Caballero and Engel (1994) try to explain the lumpy nature of investment using a model based
te
on the standard (S,s) literature. They measure imbalances in capital as the gap between the
Ac ce p
desired and the actual capital stocks (mandated investment rate). In their model, the investment rule is that once the measure of imbalance reaches a threshold value, the capital adjustment occurs at once. Firms will often wait until they reach the trigger point due to the presence of nonconvex adjustment costs. This model is empirically studied by Caballero et al. (1995) and Goolsbee and Gross (1997).
The mandated investment rate in a firm at period t, zit, is
8
αL is estimated using cost shares. The cost share of labor is calculated as the ratio of wages times employment to
the sum of the rental price of capital times capital level and wages times employment level. The employment and capital data are from COMPUSTAT database. While the rental price of capital is obtained from BLS database, the wage data are from Gray and Bartelsman's 4-digit dataset.
15 Page 15 of 60
zit ≡ where
- k-1,
(12)
and k-1 represent the natural log of the desired and the actual capital stocks in a firm at
time t, respectively. It should be noted that this measure is calculated for each firm and period;
ip t
time and firm subscripts are removed to simplify the equations. While the positive values of zit correspond to capital shortages, the negative ones indicate excess capital. Desired capital refers
cr
to the stock of capital that a firm would hold if adjustment costs were momentarily removed. It is
=
+ d,
:
an
proportional to the log of the frictionless capital stock,
us
constructed under the following assumptions. First, it is assumed that desired capital is
(13)
hold if it had never faced adjustment costs.
M
where d is a firm-specific constant.9 Frictionless capital is the stock of capital that a firm would
d
The second assumption is that the frictionless stock of capital, , is determined by a
te
neoclassical model. In this model, all frictions are assumed to be absent, including time-to-build
Ac ce p
assumption, or any adjustment costs. Y represents the value of output of an individual firm, assuming imperfect competition, and fixed factors other than capital producing decreasing returns:
(14)
where A, K, and L are profitability shocks, capital stock, and flexible factors, respectively. the capital share of production and
is
represents the flexible factor.
Optimizing over flexible factors yields a profit function: (15) 9
Bertola and Caballero (1994) show that this assumption is compatible with the behavior of a rational firm whose
profit function is isoelastic, and that faces shocks that have independent increments.
16 Page 16 of 60
where
is the price of flexible factors. Given the equation, frictionless capital is defined as (16)
ip t
where c is the cost of capital. After some manipulations and taking the logarithm of the above
(18)
cr
expression, the frictionless capital level is expressed as
where
, and y is the log of real output, which is the sum of the real
is a decreasing function of the curvature of the profit function with respect to
an
of real capital.
us
sale value of goods plus changes in the real value of finished goods inventories, and k is the log
capital, which is approximately equal to 1/(1-α) where α is the cost share of capital, estimated at , where
is the real interest
M
the 4-digit industry level.10 The cost of capital, c, is
rate, which is equal to the average nominal Baa corporate bond rate minus the measure of
d
expected inflation from the Livingston Survey of twelve-month inflation expectations.
is the
te
depreciation rate taken from the 2-digit unofficial BLS data set. In this set, depreciation rates are
Ac ce p
given for three asset groups for each 2-digit industry. The wealth share of assets is used as a weight to calculate the average depreciation rate in each 2-digit industry group. capital expenditure deflator and
is shipments price deflator taken from Gray and Bartelsman
data set (at the 4-digit industry level). The tax parameter,
where and 10
is the new
is the corporate income tax,
, is taken from the BLS database: (19)
is the present value of $1 of tax depreciation allowances,
is the effective rate of investment tax credit. This equation is given for 93 assets in each
= (rental price of capital*capital stock)/total cost of production. It is calculated at the firm-level. The data source
for rental price is BLS database and for total cost of production is COMPUSTAT. Its average value is 1.37.
17 Page 17 of 60
two-digit industry level. In order to find the two-digit weighted average values, the wealth share of assets in each industry is calculated. The average value of c is 0.16.
(20)
is the long-run elasticity of capital with respect to its cost in a firm at time t.
cr
where
ip t
With slight modifications, the empirical equation for the firm-level gap measure is:
The third and last assumption in calculating mandated investment is the estimation of
from
us
a cointegrating regression of the natural log of the capital-to-output ratio on the cost of capital at the 2-digit industry level using firm-level panel data.11 This coefficient can be interpreted as the
an
long-run elasticity of capital with respect to its cost. Bayraktar (2002a) calculates the average
M
value of this measure as -0.68. This value is close to -1, which is the long-run elasticity in neoclassical investment models.
d
After calculating frictionless capital, the firm-specific constant, d in Equation (13), is and
for the five points with investment
te
estimated by taking the average gap between
Ac ce p
closest to median investment, which can be thought of as maintenance investment. 4.2.3 Tobin’s q
Tobin's q, the ratio of the market value of firms to the replacement value of capital, is the most commonly used fundamental determinant of investment in the literature. In this study, Tobin's q is calculated following Barnett and Sakellaris (1998).12 The numerator is the sum of the market value of common stock, the liquidating value of preferred stock, the market value of long-term
11
The twice-lagged first difference of the log of cost of capital is included as explanatory variable in order to reduce
the small sample bias. The data source is COMPUSTAT. 12
Details on the calculation of this fundamental are given in Appendix B.
18 Page 18 of 60
debt and the book value of short-term debt. The denominator is the sum of the replacement value of fixed capital and inventories.
ip t
5 Empirical Results
cr
The main aim of this section is to investigate empirically systematic differences between financially constrained and unconstrained firms in their investment sensitivity to fundamentals.
us
In order to accomplish this purpose, different investment regression specifications are run first
an
for constrained versus unconstrained firms as defined below. Sensitivities of investment to fundamentals are measured by the estimated coefficients of these investment regressions. Then,
M
as reported in section 5.5, the systematic differences in investment responses to the different fundamentals are calculated across firm groups by using the estimated coefficients and shocks to
d
the fundamentals as observed in actual data.
te
In this section the following reduced form investment equation is estimated by a least
Ac ce p
squares regression technique for panel data:
(21)
where i represents firms and t represents years.13
is the investment rate at firm i in period t.
is a fundamental measure capturing investment opportunities: two types of profitability shocks (rait and ait), the mandated investment rate (zit), or Tobin's q.14 Since Bayraktar (2002a) shows that since the relationship between the investment rate and these fundamentals is 13
For the purpose of comparison, empirical specifications only with the linear term of the fundamentals are also run
and the results are reported in the regression tables. 14
It should be noted that despite the fact that Tobin’s q possibly involves measurement errors as explained in the
previous sections, it is still included in the regression specifications for the purpose of comparison.
19 Page 19 of 60
nonlinear, the squared term of the fundamentals is also included.15 FV controls for financial variables: the cash flow to capital ratio, the ratio of sales to capital, or working capital to capital.16 T represents a set of time dummies, and F a set of firm dummies to remove fixed
ip t
effects. u is the error term.
A regression specification with FV as the only independent variable is also run to understand
cr
the impact of FV on investment when fundamentals are excluded. This third specification can
us
give better information on how estimated coefficients of FV change when fundamentals are added to the empirical equation. Given that both x and FV are expected to be strong determinants
an
of investment, their joint effect can also have a significant effect on investment. In a fourth specification, an interaction term between FV and x is added in Equation (21) to analyze whether
M
or not the effect of x on investment depends on FV.
d
Each empirical specification is run separately for financially constrained versus
te
unconstrained firms. In order to identify financially constrained firms as defined in Section 4.1, the firms in the data set are split into sub panels using six alternative a priori criteria and the KZ
15
Ac ce p
index of equity dependence of firms as explained in Section 4.1.17
Related to the nonlinear relationship between fundamentals and investment, also see Barnett and Sakellaris
(1998), Barnett and Sakellaris (1999), and Abel and Eberly (2002). 16
All these ratios are calculated using the book values. The results do not change when the real values are used
instead. In addition to the financial variables mentioned above, the ratio of cash and equivalence to capital and the ratio of tax payments to income are also included. The results are not reported here because the ones with the cash and equivalence to capital ratio were similar to the results with the ratio of cash flow to capital, and the statistical significance of tax payments was negligible. 17
Other criteria such as total assets, real capital stock, flow and stock of debt burdens, and interest coverage ratio are
also introduced. But since they produce similar results, they are not reported in the paper.
20 Page 20 of 60
In subsection 5.5 the joint effects of the linear and the squared terms of fundamentals on investment are systematically investigated. The hypotheses for the results presented in that subsection are:
ip t
H0: investment/fundamental sensitivities are higher for financially constrained firms than sensitivities of unconstrained firms;
cr
H1: investment/fundamental sensitivities are not higher for financially constrained firms than
us
sensitivities of unconstrained firms.
The regression specifications used in the analysis also allow us to test the sensitivity of
an
investment to financial variables as well as the sensitivity of investment to the linear and squared terms of fundamentals separately across different firm classifications. In these tests, the statistical
M
significance of the differences in the estimated coefficients of financial variables across firm
d
groups and the differences in the estimated coefficients of the linear and the squared terms of
te
fundamentals across firm groups are calculated. The test results are reported in Tables 2 to 8. As
are:
Ac ce p
the coefficients are defined in equation (21), the null and alternative hypotheses of these tests
H0: b1 for constrained firms – b1 for unconstrained firms = 0; H1: b1 for constrained firms – b1 for unconstrained firms ≠ 0; and
H0: b2 for constrained firms – b2 for unconstrained firms = 0; H1: b2 for constrained firms – b2 for unconstrained firms ≠ 0; and H0: b3 for constrained firms – b3 for unconstrained firms = 0; H1: b3 for constrained firms – b3 for unconstrained firms ≠ 0.
21 Page 21 of 60
In the following subsections, the results are presented where firms are grouped according to: 1) the size of their capital stocks and the number of employees; 2) the level of dividend ratios; 3) the level of their debt stock and the availability of bond rating; and 4) the KZ index. After these
ip t
analyses, in the last subsection, the joint effects of linear and nonlinear terms of the fundamentals on investment are reported to summarize the systematic differences in investment/fundamental
us
cr
sensitivities between financially constrained and unconstrained firms.
5.1 Size of Firms
an
The first two criteria to classify financially constrained and unconstrained firms are the size of firms in terms of their capital stocks and the number of employees. Small firms (Class 1) are
M
defined as those with the average capital stock (or number of employees) in the bottom half of the empirical distribution. Since firms with a low capital stock typically have fewer employees,
d
the results with these two different classifications are very similar. One would expect smaller
te
firms to be financially constrained, since costs of getting external funds are presumably higher
Ac ce p
for them, and public information about their investment projects is generally more limited. These facts restrict their ability to find external funds. Given that small firms are expected to be financially constrained, their investment should respond more to changes in fundamentals or investment opportunities, because improving fundamentals indicate an improving financial position through declining agency costs. The average values of the variables for these two classifications are reported in Columns (3) to (6) in Table 1. The average investment rates in groups are similar. Smaller firms grow faster, retain a higher fraction of their income, and pay fewer dividends as a proportion of their income. When the average values of the fundamentals are compared, they are very close for the profitability shocks and the mandated investment rate. Table 1 also reports that the correlation 22 Page 22 of 60
between profitability shocks and investment is higher for small firms; they are very similar when zit is the fundamental measure and substantially low for small firms when Tobin’s q is the fundamental measure.
ip t
The estimation results based on two classifications are presented in Table 2 (size of capital stock) and Table 3 (number of employees). In almost every case, while the coefficient of the
cr
linear term of the fundamentals is larger for small firms, the coefficient of the squared term is
us
higher for large firms. The difference between the estimated coefficients of the two groups of firms is statistically significant as reported in Tables 2 and 3. The results are robust whether the
an
sample is split by the size of capital stock or by the number of employees. One possible explanation of the findings would be that smaller firms may have lower non-convex capital
M
adjustment costs, such as fixed costs, due to their smaller size. As a result, these firms can
d
linearly follow investment opportunities. Another possible explanation is that the higher
te
sensitivity of smaller firms’ investment to the shocks to the fundamentals as expected by the contracting models of financial market imperfections.18 Firms with financial constraints are more
Ac ce p
sensitive to the fundamentals (profitability shocks or mandated investment), since these fundamentals signal investment opportunities, as well as changes in the financial position. Another interesting observation is that when the relative magnitudes of the estimated coefficients are compared across class 1 and class 2 firms in regression specifications where fundamentals and financial variables are introduced together, smaller firms' investment tends to be less sensitive to changes in financial variables, even though the opposite is expected
18
The joint effects are reported in Section 5.5.
23 Page 23 of 60
according to the mainstream financial market imperfection literature.19 The test results presented in Tables 2 and 3 confirm these observations. The exception is Tobin's q, for which the coefficients of financial variables are higher for smaller firms, as presented in many empirical
ip t
studies.
Another result is that in the investment regression specifications, where only financial
cr
variables are included as independent variables, it can be seen that the magnitude of the
us
estimated coefficient of the financial variable is larger for smaller firms. But the size of the coefficient drops sharply for this group of firms, and even becomes insignificant in some
19
M
an
regression specifications, when fundamentals are introduced in the empirical model.
The high sensitivity of investment to the cash flow ratio for financially unconstrained firms is also observed in the
d
study by Kaplan and Zingales (1997). The least constrained and the most financially successful firms in their sample
te
seem to depend primarily on their cash flow to finance their investment despite the availability of additional low
Ac ce p
cost funds. In their study, they use Tobin's q to control investment opportunities, and classify firms as financially constrained by undertaking an in-depth analysis of firms. Erikson and Whited (2000) also show that cash flow does not matter, even for financially constrained firms once measurement errors are corrected. Baker, Stein, and Wurgler (2003) also report the higher sensitivity of investment to fundamentals and financial variables for highly equitydependent firms, which are identified following the methodology suggested by Kaplan and Zingales (1997). The main differences between their and this paper is that, in their analysis they do not include any nonlinear terms of fundamentals which are important determinants of investment behaviors of firms. While they use different investment measures but only one fundamental variable (Q) and one financial variable (cash flow-to-assets ratio), four different measures of fundamentals and three different financial variables are tested in this paper. Moreover, they use only the KZ index to identify financially constrained firms, while seven different classifications, including the KZ index, are used in this paper. Tsoukalas (2011) shows that time-to-build for capital projects create an investment-cash-flow sensitivity that may not be indicative of capital market frictions.
24 Page 24 of 60
There can be different reasons for why more financially constrained firms exhibit lower investment-internal funds sensitivity compared to unconstrained ones. As specified by Kaplan and Zingales (1997 and 2000), one possible reason could be excessive conservatism by
ip t
managers, which may be caused by the way firms are organized or by non-optimizating behavior of managers. Alternatively, as predicted by the contracting models, improvements in
cr
fundamentals would be more important for the investment decision of financially constrained
us
firms when compared to any improvements in internal funds alone. This is because their financial constraints start relaxing more with good fundamental shocks due to a declining agency
an
cost of borrowing.
Given the statistical and economic significance of fundamentals and financial variables in
M
determining fixed investment, a regression with an interactive term between financial variables
d
and fundamentals is included to understand their joint impact on investment. In Tables 2 and 3 it
te
can be seen that, in general, the coefficients of the interaction term are positive and statistically significant for smaller firms, while it is negative and significant or positive but insignificant for
Ac ce p
larger firms. The trend is robust across different financial variables, fundamentals, and firm classifications. The positive and significant coefficient of the interactive term for smaller firms means that at higher values of fundamentals, the impact of financial variables on investment gets larger; or at higher values of financial variables, the impact of fundamentals on investment gets larger. As a result, for smaller firms the effect of financial variables on investment positively depends on fundamentals, and at the same time, it suggests that the effect of fundamentals on investment depends on financial variables. Given that smaller firms tend to be more financially constrained, the positive and significant interaction term reinforces the argument that the joint impact of the two determinants of investment (fundamental and financial) are essential in the
25 Page 25 of 60
investment process of financially constrained firms. On the other hand, for financially less constrained firms the coefficient of the interaction term is negative, or positive but not significant. For negative and significant coefficients, it can be concluded that at higher values of
ip t
fundamentals the impact of financial variables on investment gets smaller for less constrained firms. In this case, the effect of fundamentals on investment depends negatively on financial
cr
variables. The two determinants of investment do not reinforce each other in their fixed
us
investment activities. Given that these firms are expected to be less financially constrained, it is
an
an expected result.
5.2 Size of Dividends
M
The third and the fourth criteria used in identifying financially constrained firms are the dividend payout ratio and the dividend to capital ratio, respectively. Firms are classified as financially
d
constrained (Class 1) if their ratios are in the bottom half of the empirical distribution. Class 2
te
represents groups with higher dividend payments. One would expect Class 2 to be financially
Ac ce p
unconstrained since they pay a large fraction of their income as dividends, which suggests that the opportunity cost of their internal funds is low (Fazzari et al., 1988). The average values of the variables for all groups are reported in Columns (7)-(10) in Table 1. The capital stock and the number of employees are lower for Class 1 firms; but they grow faster and retain a larger fraction of their income. While the average investment rate is 0.15 for Class 1 firms, it is 0.11 in Class 2. The simple correlations between investment and fundamentals indicate that the sensitivity of investment for high dividend paying firms to the fundamentals is stronger. One explanation for this unexpected result can be that high-dividend paying firms would actually be more dependent on fundamentals to be able to more easily finance their
26 Page 26 of 60
investment, given that they distribute a higher proportion of their internal funds as dividends. In this case, it looks like higher-dividend payers are more financially constrained. However, this observation which is based on a simple correlation analysis changes with
ip t
regression analysis. The estimation results are presented in Tables 4 and 5. In reviewing Table 4, in general the sensitivity of investment to the linear term of the fundamentals is higher for the
cr
firms with low dividend payouts, while the sensitivity to the squared term is higher for the firms
us
with high dividend payouts. As presented in the tables, the difference between the estimated coefficients of the two groups of firms is statistically significant in most cases. There are some
an
exceptions to this trend. When rait is the fundamental measure of investment and the sales-tocapital ratio or working capital-to-capital ratio are the financial variables, the linear term of
M
fundamentals is smaller for the financially constrained group. However, in these cases the
d
difference between the coefficients of the two groups of firms is not statistically significant.
te
As was the case with smaller firms, those with lower dividend payouts seem to follow fundamentals linearly. Again the results support the expectations of contracting models. Given
Ac ce p
that the magnitude of the nonlinear term in some results is much higher for the financially unconstrained firms, the joint effect of linear and nonlinear terms of the fundamentals on investment is expected in some cases to be close for the financially constrained and unconstrained groups. This is further discussed in Section 5.5. In Table 5, we observe the trend that financially constrained firms’ investment responds more to the linear term of fundamentals, but less to the squared term, only when zit is the fundamental measure. When ait is the fundamental determinant of investment, the size of both linear and squared terms is larger for financially constrained firms. As a result, ait is predicted to have a stronger impact on investment when firms are classified based on the level of dividends-to-
27 Page 27 of 60
capital ratio. For rait and q, the results are mixed. It is worth noting that when rait is the fundamental determinant, the difference between the estimated coefficients of Class 1 firms and Class 2 firms is not statistically significant in most cases.
ip t
Even though Class 1 firms are expected to be financially constrained, the sensitivity of the investment rate to the cash flow ratio across firm classifications is lower compared to its
cr
sensitivity to the fundamentals, except in the case of Tobin’s q, which may have a possible
us
mismeasurement problem. As was the case in Tables 2 and 3, for expected-to-be financially constrained firms, the coefficients of the fundamentals are robust to the inclusion of financial
an
variables in the regression, while the coefficients of the financial variables drop significantly when they are included together with fundamentals in the same equation. The results also show
M
that the interaction terms between financial variables and fundamentals again tend to have a
d
positive and significant effect for the constrained group, while it usually has a negative and
te
significant or positive but insignificant effect on the unconstrained group.
Ac ce p
5.3 Size of debt to capital ratio and bond rating The last two criteria used in identifying financially constrained firms are the availability of a bond rating and the size of the debt to capital ratio. Since the debt burden of firms with a high debt ratio or without any bond rating (Class 1) is heavier, they are expected to be more financially constrained. Therefore, their investment should be more sensitive to changes in internal funds as expected by the mainstream financial market imperfections literature, or alternatively it is expected to be more sensitive to the fundamentals as expected by the contracting models of financial frictions. Firms with a bond rating are expected to be less financially constrained because public debt issuance is a good indicator of firms' financial position, given that it provides a low-cost access to capital markets (Whited, 1992; Calomiris et 28 Page 28 of 60
al., 1994; Agca and Mozumdar, 2008). Firms with a higher debt burden are expected to have more financial problems because they are expected to pay higher premiums to obtain external funds (Bernanke et al., 1990; Whited, 1992; Hu and Schiantarelli, 1998; Hennessy et al., 2007).
ip t
Columns (11)-(14) in Table 1 show the average values of the variables for the firms classified based on these two criteria. The share of firms with a bond rating is 0.30. The capital
cr
stock and the number of employees of firms without a bond rating are lower. Firms with a low
us
debt to capital ratio have a larger amount of capital on average, but they have fewer employees. The average investment rate of firms with a high debt to capital ratio is slightly higher. The
an
simple correlation between investment and the fundamentals tends to be stronger for the financially constrained firms except in the case of Tobin’s q.
M
Tables 6 and 7 exhibit the estimation results. In Table 6 the investment rate of firms without
d
a bond rating or with a high debt to capital ratio tends to respond more to changes in the linear
te
term of the fundamentals, but less to the squared term, as was the case with other classifications. This observation is true even in the specifications with Tobin’s q. For constrained firms (firms
Ac ce p
without a bond rating), the dropping economic significance of financial variables in determining investment in the presence of fundamentals can supports the argument that fundamentals can capture impacts of financial variables on investment. Table 7 shows that the sensitivity of the investment rate to both financial variables and fundamentals is stronger for the firms with a high debt to capital ratio (Class 1) when compared to the sensitivity of Class 2 firms. The exception is that when zit is the fundamental determinant of investment, both linear and squared terms of fundamentals are statistically significantly higher for constrained firms. It is important to note that the size of the financial variable tends to be higher for Class 1 firms. This is an expected result since these firms have a heavier debt burden.
29 Page 29 of 60
Consequently, any changes in financial variables or fundamentals should have a larger effect on the investment rate. In Table 7, when the cash-flow ratio is the financial variable in the regression, the explanatory power of this financial variable decreases for the firms with a high
ip t
debt ratio relative to the explanatory power of the same financial variable for the firms with lower debt burden. However, the significance of the fundamentals continues in the same
cr
equations for Class 1 firms.
us
As was the case in other classifications, it is observed that the estimated coefficients of the interaction terms between financial variables and fundamentals are positive and significant for
an
constrained firms, while they are mostly negative or insignificant for firms with bond rating or
M
lower debt burden.
d
5.4 Firm Classification Based on the KZ index
te
The KZ index is initially introduced by Kaplan and Zingales (1997) to identify financially constrained firms. They name potentially financially constrained firms as equity-dependent
Ac ce p
firms. Using this classification, Kaplan and Zingales (1997) show that equity-dependent firms are more sensitive to fundamentals. While constructing this index, Kaplan and Zingales undertook a detailed study of the financial constraints of 49 low-dividend manufacturing firms. Based on different subjective and objective criteria, they sorted these firms from least to most constrained. After creating this ranking, they ran a regression to link this new index to the series from COMPUSTAT so that the index could be used in other studies. Lamont, Polk, and SaaRequejo (2001) and Baker, Stein, and Wurgler (2003) are two prominent examples of empirical studies which construct the KZ index using the COMPUSTAT database. In this paper, following
30 Page 30 of 60
the methodology of Baker, Stein, and Wurgler (2003), the KZ index is created using the
where KZ is the index;
is the ratio of cash flow (items 14 and 18 in COMPUSTAT) to
is the ratio of dividends (items 21 and 19) to assets;
is the ratio of cash
cr
assets;
ip t
following equation in which the estimated coefficients come from Kaplan and Zingales (1997):
is leverage (items 9+34 divided by items 9+34+216). The
us
balances (item 1) to assets; and
an
higher the index value, the higher equity dependence is. When the firms are ranked based on their average index value from low to high and split into two groups, firms in the high value
M
group are expected to be equity dependent or financially constrained (Class 1), and firms with lower index values are considered to be financially unconstrained (Class 2).
d
Table 1 (columns (15) and (16)) reports the descriptive statistics for these two groups of
te
firms. Equity-dependent firms tend to be smaller both in terms of their capital stock and the
Ac ce p
number of employees. They tend to invest slightly less. While their dividend payout is lower, their debt ratios are higher. The correlation between the investment rate and the fundamentals are higher for equity-dependent firms, except when q is the fundamental determinant of investment. Table 8 reports the estimated coefficients for different specifications and firm classifications. The results support the previous findings. For equity-dependent firms, which are expected to be financially constrained, both the linear and squared terms of the fundamentals are higher when rait and q are the fundamental determinants. However, the squared term is lower for them when the fundamentals are measured by ait and zit. In section 5.5, the joint effects of the linear and squared terms of the fundamentals are systematically investigated.
31 Page 31 of 60
In general, the estimated coefficients of the financial variables are statistically significantly higher for equity-dependent firms than the ones for equity-independent firms. The sign, significance, and the magnitude of the interaction terms between the financial variables and
cr
ip t
fundamentals indicate a strong, positive joint effect on investment for Class 1 firms.
5.5 Joint Effects of Linear and Nonlinear Terms of Fundamentals on Investment
us
For the most part, regression results indicate that the coefficient of the linear term of the fundamentals is higher for the financially constrained firms, while the coefficient of the squared
an
term is higher for the financially unconstrained firms. These findings are robust across
M
alternative classifications used in distinguishing the financially constrained firms and across alternative methods to calculate the fundamentals. Before any conclusion is made, it is important
d
to assess for which group of firms the joint effects of the linear and nonlinear terms are stronger
te
in the sample used for estimation. In this assessment the estimated coefficients of the linear and
used.
Ac ce p
squared terms of the fundamentals and the standard deviations of the fundamental shocks are
For these analyses, different shock values of each fundamental are calculated and reported in the first row of Table 9. The shock values are within ±1 or ±2 standard deviations of the shocks. A set of estimated coefficients are needed to calculate the joint effects. Given that the results are mostly robust to which financial variable is included in the specifications, and given that the cash-flow to capital ratio (CF_K) is one of the most commonly used financial variables in the literature, the following regression equation is picked to construct the joint effects of the linear and nonlinear terms of the fundamental shocks: (22) 32 Page 32 of 60
where x stands for the fundamentals ait, rait, or zit. The estimated coefficients of this specification are reported in Columns (6) and (10) of Tables 2 to 8. While the results in Column (6) are for the financially constrained firms, the ones in Column (10) are for the unconstrained where
ip t
firms. The values in Table 9 are calculated as:
coefficients of the linear and nonlinear terms of the fundamentals.
cr
the shock values are reported in the first row of the table and b1 and b2 are the estimated
us
The joint effects of the linear and the squared terms of the fundamentals are substantially stronger for the financially constrained firms within the ±2 standard deviation limits of the
an
shocks. The asymmetry between the effects of the negative versus positive shocks is substantial. The firms in the financially constrained groups respond more strongly to the positive shocks
M
when compared to the responses by the unconstrained firms. The financially constrained firms
d
react even more strongly to the negative shocks by cutting their investment significantly. There
te
are only a few exceptions where the response of investment to the positive shocks is slightly higher for the unconstrained firms. All of these exceptional cases belong to the classifications
Ac ce p
based on dividend payouts or the ratio of dividends to capital. Overall, when a realistic size of a shock is assumed, the estimated linear and nonlinear coefficients of the fundamentals imply that financially constrained firms invest more when a positive shock realizes, and cut their investment more significantly when a negative shock occurs.
6 Conclusion The aim of this paper is to investigate whether or not the empirical and economic significance of fundamentals, measured by profitability shocks and mandated investment rate, are higher in determining fixed investment behavior of financially constrained firms, as is predicted by the 33 Page 33 of 60
contracting models of financial frictions. The main point is that fundamentals are indicators of improving investment opportunities, as well as easing financial constraints through dropping agency costs of borrowing. Therefore, investment by financially constrained firms is expected to
ip t
respond more strongly to changes in fundamental determinants of investment. The findings based on regression analyses support this expectation. One common finding is that the
cr
coefficient of the linear term of the fundamentals in investment regressions tends to be higher for
us
financially constrained firms, while the coefficient for the nonlinear term of the fundamentals tends to be higher for financially unconstrained firms. But the joint effects of the linear and
an
nonlinear terms of the fundamentals within ±2 standard deviation of the shocks are stronger for financially constrained firms whether shocks are positive or negative. These results are quite
M
robust across different methods to construct the fundamentals and different classification
d
methods to identify the financially constrained firms. This would be a useful finding for
its critics.
te
researchers since it cannot be easily explained by the framework of Fazzari et al. (1988), nor by
Ac ce p
The results show that financial variables are still statistically significant determinants of firmlevel fixed investment. But the interesting outcome is that in most cases the explanatory power of the financial variables in the investment process tends to be lower for expected-to be financially constrained firms when compared to the explanatory power of the same financial variables for the unconstrained group of firms. This finding further supports the argument that constrained firms are more dependent on the changes in fundamentals to ease their financial burdens, as expected by contracting models. For the purpose of comparison, the same analyses are repeated using Tobin's q as a measure of fundamental. In regression specifications, in which Tobin’s q is the fundamental, financial
34 Page 34 of 60
variables are relatively more significant in explaining investment compared to the impact of Tobin’s q (fundamental measure) on investment. The result is robust for both constrained and unconstrained firms. These outcomes indicate that the statistical and economic significance of
the inadequacy of Tobin's q in capturing investment opportunities.
ip t
financial variables, but not fundamentals, in explaining investment indeed would be caused by
cr
In terms of future studies related to this topic, the robustness of the results across different
us
industries should be investigated. One possibility is that the analyses may focus only on durable-
Ac ce p
te
d
M
an
goods industries, which are more homogenous, compared to nondurables industries.
35 Page 35 of 60
References Abel, A.B., Eberly, J.C., 2002. Investment and q with Fixed Costs: an Empirical Analysis. Unpublished results, University of Pennsylvania (January).
ip t
Abel, A.B., Eberly, J.C., 2003. Q Theory Without Adjustment Costs and Cash Flow Effects
cr
Without Financing Constraints. Unpublished results, University of Pennsylvania (October).
us
Agca, S., Mozumdar, A., 2008. The impact of capital market imperfections on investment–cash
an
flow sensitivity. Journal of Banking & Finance 32, 207–216.
Albuguerque, R., Hopenhayn, H., 2004. Optimal Lending Contracts and Firm Dynamics. Review
M
of Economic Studies 71(2), 285-315.
Baker, M., Stein, J. C., Wurgler, J., 2003. When Does the Market Matter? Stock Prices and the
te
969-1005.
d
Investment of Equity-Dependent Firms. The Quarterly Journal of Economics 118(3),
Ac ce p
Barnett, S., Sakellaris, P., 1998. Nonlinear Response of Firm Investment to Q: Testing a Model of Convex and Non-convex Adjustment Costs. Journal of Monetary Economics 42, 261288.
Barnett, S., Sakellaris, P., 1999. A New Look at Firm Market Value, Investment and Adjustment Cost. The Review of Economics and Statistics 81, 250-60. Bayraktar, N., 2009. Investment, Alternative Measures of Fundamentals, and Revenue Indicators. International Journal of Revenue Management 3(2), 148-178. Bayraktar, N., 2002a. Analyses of Alternative Fundamentals of Fixed Capital Investment. Unpublished results, University of Maryland.
36 Page 36 of 60
Bayraktar, N., 2002b. Effects of Financial Market Imperfections and Non-convex Adjustment Costs in the Capital Adjustment process. Unpublished results, University of Maryland.
Investment. ECB Working Paper Series No. 566 / December 2005.
ip t
Bayraktar, N., Sakellaris, P., Vermeulen, P., 2005. Real versus Financial Frictions to Capital
cr
Bernanke, B.S., Campbell, J.Y., 1988. Is there a Corporate Debt Crisis?. Brookings Papers on Economic Activity 1, 83-139.
us
Bernanke, B.S., Campbell, J.Y., Whited, T., 1990. US Corporate Leverage: Developments in
an
1987 and 1988. Brookings Papers on Economic Activity 1, 255-286. Bernanke, B.S.; Gertler, M., 1990. Financial Fragility and Economic Performance. Quarterly
M
Journal of Economics (February), 87-114.
Bernanke, B.S., Gertler, M., Gilchrist, S., 2000. The Financial Accelerator in a Quantitative
d
Business Cycle Framework. In Taylor, J.B., Woodford, M. (Eds.), Handbook of
te
Macroeconomics, Vol.3, North Holland.
Ac ce p
Bertola, G., Caballero, R.J., 1994. Irreversibility and Aggregate Investment. Review of Economic Studies 61, 223-246. Brainard, W.C., Shoven, J.B., Weiss, L., 1980. The Financial Valuation of the Return to Capital. Brookings Papers on Economic Activity 2, 453-511. Caballero, R.J., Engel, E.M.R.A., 1994. Explaining Investment Dynamics in the U.S. Manufacturing : A Generalized (S,s) Approach. NBER Working Paper No. 4887. Caballero, R.J., Engel, E.M.R.A., Haltiwanger, J., 1995. Plant-Level Adjustment and Aggregate Investment Dynamics. Brookings Papers on Economic Activity 2, 1-39.
37 Page 37 of 60
Calomiris, C.W., Himmelberg, C.P., Wachtel, P., 1994. Commercial Paper, Corporate Finance, and The Business Cycle: A Microeconomic Perspective. NBER Working Paper, No. 4848 (September).
ip t
Carlstrom, C., Fuerst, T., 1997. Agency Costs, Net Worth, and Business Fluctuations: A
cr
Computable General Equilibrium Analysis. American Economic Review 87, 893-910. Chatterjee, S., Corbae, D., Nakajima, M., Rios-Rull, J-V., 2007. A Quantitative Theory of
us
Unsecured Consumer Credit with Risk of Default. Econometrica 75(6), 1525-89.
an
Clementi, G.L., Hopenhayn, H., 2006. A Theory of Financing Constraint and Firm Dynamics. Quarterly Journal of Economics 121(1), 229-265.
M
Cooley, T., Marimon, R., Quadrini, V., 2004. Aggregate Consequences of Limited Contract Enforceability. Journal of Policy Economics 112, 817-847.
te
Review 91(5), 1286-1310.
d
Cooley, T., Quadrini, V., 2001. Financial Markets and Firm Dynamics. The American Economic
Ac ce p
Cooper, R.W., Ejarque, J., 2001. Exhuming Q: market power vs. capital market imperfections. NBER Working Paper Series; No. 8182 (March). Cooper, R.W., Haltiwanger, J.C., 2006. On the Nature of Capital Adjustment Costs. Review of Economic Studies 73(3), 611-33. Del Boca, A., Galeotti, M., Rota, P., 2008. Non-convexities in the adjustment of different capital inputs: A firm-level investigation. European Economic Review 52, 315–337. Erickson, T., Whited, T.M., 2000. Measurement Error and the Relationship between Investment and q. Journal of Political Economy 108(5), 1027-57.
38 Page 38 of 60
Fazzari, S.M., Hubbard, G.R., Petersen, B.C., 1988. Financing Constraints and Corporate Investment. Brookings Paper for Economic Activity 1, 141-195. Gilchrist, S., Himmelberg, C.P., 1995. Evidence on the Role of Cash Flow for Investment.
ip t
Journal of Monetary Economics 36, 541-572.
cr
Gilchrist, S., Himmelberg, C.P., 1998. Investment, Fundamentals and Finance. NBER Working Paper # 6652 (July).
us
Gomes, J.F., 2001. Financing Investment. American Economic Review 91(5), 1263-85.
an
Goolsbee, A., Gross, D.B., 1997. Estimating Adjustment Costs with Data on Heterogeneous Capital Goods. NBER WP No. 6342 (December).
M
Hennessy, C.A., Levy, A., Whited, T.M., 2007. Testing Q theory with financing frictions. Journal of Financial Economics 83, 691–717.
d
Hennessy, C.A., Whited, T.M., 2007. How costly is external financing? Evidence from a
te
structural estimation. The Journal of Finance 62(4), 1705–1745.
Ac ce p
Hu, X., Schiantarelli, F., 1998. Investment and Capital Market Imperfections: A Switching Regression Approach Using U.S. Firm Panel Data. The Review of Economics and Statistics 80(3), 466-479.
Kaplan, S.N., Zingales, L., 1995. Do Financing Constraints Explain Why Investment is Correlated with Cash Flow?. NBER Working Paper. No. 5267 (September). Kaplan, S.N., Zingales, L., 1997. Do Investment Cash Flow Sensitive Provide Useful Measure of Financing Constraints?. Quarterly Journal of Economics 112(1), 169-215. Kaplan, S.N., Zingales, L., 2000. Investment-Cash Flow Sensitivities are not Valid Measures of Financing Constraints. NBER Working Paper. No. 7659 (April).
39 Page 39 of 60
Lamont, O., Polk, C., Saa-Requejo, J., 2001. Financial Constraints and Stock Returns. Review of Financial Studies XIV, 529-554. Marcet, A., Marimon, R., 1992. Communication, Commitment and Growth. Journal of Economic
ip t
Theory 58(2), 219-249.
cr
Salinger, M., Summers, L.H., 1983. Tax Reform and Corporate Investment: A Microeconomic Simulation Study. In: Martin, F. (Eds.) Behavioral Simulation Methods in Tax Policy
us
Analysis.
Journal of Economic Theory 21, 265-293.
an
Townsend, R., 1978. Optimal Contracts and Competitive Markets with Costly State Verification.
M
Tsoukalas, J.D., 2011. Time to build capital: Revisiting investment-cash-flow sensitivities. Journal of Economic Dynamics & Control 35, 1000–1016.
d
Whited, T.M., 1992. Debt, Liquidity Constraints, and Corporate Investment: Evidence from
Ac ce p
te
Panel Data. Journal of Finance 47(4), 1425-1460.
40 Page 40 of 60
APPENDIX A Sample Selection
Ac ce p
te
d
M
an
us
cr
ip t
The major data source is Standard & Poor's COMPUSTAT Database, which is a database of financial, statistical, and market information covering publicly traded companies. The data set covers only U.S. manufacturing firms with a SIC code between 2000-3999 for the years from 1981 to 1996. The calendar year, which is a period of one year beginning with January 1 and ending with December 31, is taken. In the initial data set, the number of U.S. manufacturing firms was 3798 active firms and 3503 inactive firms, for which data are no longer collected because of such reasons as merger, acquisition, and bankruptcy. The elimination of firms is conducted following several steps: 1. I filter the data depending on the availability of the plant, property, and equipment (PPE) series. I delete firms with missing data points more than two periods for the capital expenditure for PPE and for total gross PPE between 1981 and 1996. All foreign firms are also eliminated. The total number of firms after this elimination is 1200. 2. Then I delete the firms, which have involved in significant acquisitions or merger activity. There are different ways to identify these firms. One way is the elimination of firms, which have expenditure on acquisitions accounted for more than 15 percent of assets. This way has not been used since the data for acquisitions are missing for many firms. The other commonly used way is the exclusion of firms whenever |Gt – Gt-1 – It + Rt| > 0.15Gt-1, where Gt is the book value of the capital stock, It is nominal capital expenditure and Rt is the retirement of capital. This method is taken from Gilchrist and Himmelberg (1995). For some firms, the retirement series are missing. In this case, the retirement value is set to be zero unless the left hand side of the formula is negative. In this case, 10 percent of the previous period's capital is taken as retirement value. The idea behind this formula comes from the accounting identity: the book value of physical capital is equal to the capital level of the previous period plus nominal investment at the current period minus retirement plus the residual balance, which includes changes in capital caused by mergers or acquisitions. So if a firm's residual balance is greater than 0.15 of its previous period capital, then it is assumed that this firm gets into a merger or acquisition activity, and then it is deleted. After applying this method, 500 firms are left. 3. Two firms are eliminated since their data are discontinuous due to FASB 94 (Financial Accounting Standards Board), which requires wholly owned subsidiaries to be included on the firm's balance sheet. The total number of firms drops to 498 after deleting these two firms. 4. The investment series is constructed through subtracting the sale of PPE and retirement of PPE series from the capital expenditure on PPE series. Due to the presence of additional missing data points for the sale of capital and retirement series, the investment rate series could not be constructed for some firms. This fact decreases the number of firms to 471. 5. The replacement value of the capital stock is constructed using a perpetual inventory method as specified in Section 4.1. 8 of firms are deleted since this method produces negative capital stock for them. After all, the number of firms is 463.
41 Page 41 of 60
APPENDIX B Details on the Calculation of Tobin’s q
ip t
The numerator of average Tobin's q is the sum of the market value of common stock, the liquidating value of preferred stock, the market value of long-term debt, and the book value of short-term debt. The denominator is the sum of replacement value of fixed capital and inventories. These variables are calculated as follows:
if INVt* INVt*1
an
PPI t INVt* INVt*1 INVt INVt 1 PPI t 1 PPI t INVt ( INVt 1 INVt* INVt*1 ) PPI t 1
us
cr
Replacement value of inventories: For firms using the first-in-first-out (FIFO) method, inventories are valued at current cost, thus the book value equals the replacement value for them. For firms using the last-in-first-out (LIFO) or any other method, inventories are valued at historic cost. While converting the book value to the replacement value, for the first year, the book value is taken equal to the replacement value. Following Salinger and Summers (1983) and Whited (1992), the following formulas are used for the following years:
if INVt* INVt*1
M
where INVt is the replacement value of LIFO inventories at time t and INVt* is their reported book value.
Ac ce p
te
d
Market Value of Long-Term Debt: The method suggested by Bernanke and Campbell (1988) and Whited (1992) is used on converting the book value of long-term debt to the replacement value. Since the COMPUSTAT database provides only limited information on maturities of debt, it is necessary to construct the maturity distribution of long-term debt from historical information on debt issues. Firstly, following Brainard, Shoven and Weiss (1980), it is assumed that all longterm debts mature in twenty years. For the first year, each individual firm's maturity distribution is set equal to the aggregate taken from Historical Statistics of the United States, series X 499509, p. 1005 for the years 1961-1970. I give equal weight to the maturity distribution for years 1971-80. Then, if Djt is debt due in j years at time t, LTDt is the reported value of long-term debt at time t, and DIt is the amount of debt issued at time t, the maturity distribution is updated as follows:
D20t DIt LTDt ( LTDt 1 D1,t 1 )
if LTDt ( LTDt 1 D1,t 1 ) 0
D20t DIt 0
if LTDt ( LTDt 1 D1,t 1 ) 0
D jt D j 1,t 1,
j = 1, …, 19.
If LTDt ( LTDt 1 D1,t 1 ) 0 , debt due in one to nineteen years is scaled down by the factor: LTDt . LTDt 1 D1,t 1
42 Page 42 of 60
The actual values of debt due in one to five year are available in COMPUSTAT. These values are replaced by the calculated values and the rest of the maturity distribution is rescaled in order to be consistent with total amount of long-term debt:
D ajt D*jt ,
j = 1, …, 5;
ip t
5 ( D jt D*jt ) , D ajt D jt 1 j 1 20 D jt j 6
cr
j = 6, …, 20;
us
where D ajt is the adjusted value of debt due in j years and D*jt , is the reported value of debt due in j years for j equal to one to five.
IEX t
M
NLTDt LTDt
an
The final modification adjusts the book value of total book value of debt to the reported interest expense consistent with that implied by assuming that the firm's interest expense at time t is the Baa rate at time t. The new value of total book value is scaled as follows:
,
20
Baa j 1
t j 20
D
a jt
te
d
where NLTDt is the scaled value of long-term debt at time t, Baat is the interest rate on grade Baa bonds at time t and IEXt is the book value of interest expense at time t. Then the new maturity distribution is set proportional to the old distribution.
Ac ce p
Market value of Equity: The value of common stock at the beginning of each year is estimated, following Salinger and Summers (1983), as the closing price of a share of stock for each company in year t-1 times the number of outstanding shares at t-1. The value of preferred stock is estimated by dividing preferred cash dividends by the Standard and Poor's preferred stock yield (taken from CITIBAS database).
43 Page 43 of 60
ip t cr
Table 1 - Average Values of Variables for Different Firm Classifications and Correlations between Fundamentals and Investment
Firms with Firms with low high number of number of employees employees
Firms with Firms with low high dividend dividend payout payout ratio ratio
1511.12
6943.51
27.71
3034.49
43.04
1880.46
8121.73
33.05
3322.61
51.23
0.13
0.34
0.13
0.12
0.14
13.40
48.95
0.86
25.24
0.66
0.15
4.24
0.21
0.07
0.22
0.68
3.86
0.81
0.56
0.90
0.39
0.65
0.41
0.35
0.30
4.11
0.22
0.04
0.18
0.21
Dividend to capital ratio (9) (10)
Firms with Firms with low high dividend dividend to capital to capital ratio ratio
Bond rating (11) (12) Firms without bond rating
Firms with bond rating
Total debt to capital ratio (13) (14) Firms with Firms with high debt low debt to capital to capital ratio ratio
KZ index (15) (16) Equitydependent firms (high constraint)
Not equitydependent firms (low constraint)
3005.50
965.70
2057.19
437.22
2592.75
232.01
4805.24
1214.08
1823.18
1246.19
1727.06
3289.48
1084.36
2228.98
498.36
2822.17
256.00
5261.34
1406.09
1922.43
1440.33
2052.53
0.12
0.15
0.11
0.14
0.11
0.13
0.12
0.14
0.12
0.12
0.13
25.25
11.28
14.58
4.61
21.21
3.08
38.07
16.07
9.80
10.49
14.54
0.07
0.24
0.05
0.23
0.06
0.17
0.08
0.21
0.08
0.19
0.22
0.47
1.04
0.34
0.91
0.47
0.69
0.67
0.76
0.61
0.81
0.57
0.41
0.35
0.46
0.30
0.48
0.29
0.38
0.39
0.60
0.16
0.54
0.25
0.40
0.11
0.51
-0.09
0.70
0.08
0.54
0.32
0.30
0.23
0.39
0.14
0.44
0.05
0.04
0.04
0.05
0.01
0.08
0.01
0.08
0.05
0.04
0.03
0.06
0.01
0.07
0.49
0.22
0.20
0.21
0.21
0.18
0.24
0.15
0.27
0.22
0.20
0.16
0.26
0.12
0.28
2.86
2.80
3.25
2.33
3.02
2.57
2.82
2.77
2.64
2.94
3.01
2.25
3.03
2.56
2.69
3.16
0.86
1.89
1.13
0.47
1.07
0.53
0.91
0.71
0.81
0.81
0.97
0.38
0.70
0.91
0.67
1.12
0.00 0.01 0.02 0.25
Correlations Whole sample 0.254 0.183 0.290 0.126
0.20
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.28
0.01
0.00
0.01
0.00
0.01
0.00
0.01
0.00
0.01
0.00
0.01
0.01
0.01
0.00
0.29
0.02
0.01
0.01
0.02
0.02
0.01
0.02
0.01
0.02
0.01
0.01
0.02
0.01
0.02
0.50
0.37
0.13
0.37
0.14
0.36
0.15
0.28
0.22
0.30
0.12
0.26
0.24
0.12
0.38
0.281 0.204 0.287 0.110
0.191 0.146 0.295 0.251
0.274 0.195 0.293 0.113
0.204 0.161 0.285 0.215
0.239 0.173 0.288 0.120
0.289 0.222 0.300 0.163
0.226 0.169 0.290 0.117
0.308 0.226 0.293 0.221
0.275 0.200 0.290 0.114
0.143 0.116 0.288 0.288
0.258 0.185 0.326 0.109
0.235 0.186 0.241 0.251
0.271 0.204 0.307 0.121
0.237 0.158 0.267 0.241
Ac
Corr(inv ,ait ) Corr(inv ,rait ) Corr(inv ,zit ) Corr(inv ,qit )
Firms with Firms with small large capital capital stock stock
ed
Fundamentals Profitability shocks (ait ) Profitability shocks (rait ) Mandated inv. rate (zit ) Log of Tobin's q
Dividend payout ratio (7) (8)
ce pt
Plant, Property, and Equipment (PPE, in millions of 1992 dollars) PPE (gross book value, in millions of dollars) Investment rate Employees (in thousands) Growth rate of real (net) sales Earnings retention rate Total Debt over PPE Dividends payout ratio Dividends over PPE Cash flow to PPE ratio Sales (net) to PPE ratio Working capital to PPE ratio
Standard deviation
Number of employees (5) (6)
us
Mean
Size of capital stock (3) (4)
M an
Whole sample (1) (2)
Note: Corr(.) is the correlation. inv is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q.
44 Page 44 of 60
ip t
Class 1: small capital stock
ait2
0.26 0.26 0.16 0.16 [17.5] [17.6] [12.3] [12.3] 0.20 0.26 [3.9] [5.3]
FV ait*FV Adj. R2 0.22 0.22 0.25 0.26 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.11 [3.88] Class 1 ait2 - Class 2 ait2 -0.07 [-2.39] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 0.11 0.12 [11.3] [11.9] [10.2] [11.1] 0.11 0.14 [4.4] [5.3]
rait*FV
(5)
(8)
0.24 0.24 [16.5] [16.6] 0.17 [3.4] 0.06 0.06 [5.5] [5.5]
0.23 [15]
0.12 0.13 [10.4] [10.9] 0.09 [3.7] 0.04 0.03 [2.5] [2.3]
0.12 [9.9]
0.38 0.06 [14] [5.6] 0.07 [2.8] 0.23 0.23 0.10 0.23 Column (6) & Column (10) 0.09 [3.66] -0.10 [-5.29] -0.07 [-4.79]
0.38 0.02 [14] [1.7] 0.05 [2.1] 0.19 0.19 0.10 0.19 Column (6) & Column (10) 0.03 [2] -0.04 [-4.99] -0.07 [-4.5]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.18 0.18 0.23 0.24 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.01 [2.74] Class 1 rait2 -Class 2 rait2 -0.03 [-0.4] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.15 0.15 0.27 [11.9] [11.8] [13] 0.27 [5.5] 0.14 0.14 0.16 0.22 [7.5] [7.7] [8.4] [10.3] -0.60 [-7.3] 0.26 0.27 0.23 0.28
0.09 0.10 0.09 [8.3] [9.2] [5.5] 0.13 [5] 0.11 0.10 0.16 0.11 [5.4] [5.1] [8.4] [5.4] 0.01 [0.2] 0.24 0.24 0.23 0.24
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: large capital stock
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.22 0.22 0.11 0.11 0.11 0.10 0.23 0.23 0.16 0.13 0.13 0.11 [13.7] [13.8] [5.2] [8.6] [8.6] [4.5] [15.6] [15.7] [8.6] [10.3] [10.3] [5.9] 0.19 0.25 0.19 0.23 [3.7] [5.1] [3.8] [4.8] 0.02 0.02 0.16 0.03 0.06 0.06 0.07 0.06 0.04 0.04 0.19 0.05 0.13 0.13 0.15 0.13 [8] [7.9] [32.8] [9] [13.7] [13.7] [17.3] [13.5] [10.2] [10.2] [19.6] [10.8] [11.4] [11.2] [13.2] [11.1] 0.03 0.00 0.06 0.04 [6.7] [0.3] [7] [1.6] 0.24 0.24 0.31 0.25 0.30 0.31 0.29 0.30 0.25 0.25 0.15 0.26 0.29 0.29 0.26 0.29 Column (14) & Column (18) Column (22) & Column (26) 0.11 [1.76] 0.10 [1.92] -0.06 [-5.29] -0.04 [-4.95] -0.04 [-4.79] -0.08 [-1.39]
M an
Financial Variables
cr
Table 2 -Response of investment to alternative fundamentals and financial variables (when firms are grouped by the size of their capital stocks)
0.07 0.08 0.11 [6.2] [6.7] [6.4] 0.07 [3.2] 0.03 0.03 0.16 0.04 [10.1] [9.6] [32.8] [9.6] 0.02 [2.9] 0.21 0.21 0.31 0.21 Column (14) & Column (18) 0.03 [1.76] -0.03 [-4.38] -0.03 [-5.5]
0.04 0.05 0.09 [3.8] [4.6] [5] 0.10 [4] 0.07 0.06 0.07 0.07 [14] [13.6] [17.3] [14.3] -0.02 [-3.3] 0.28 0.29 0.29 0.29
0.09 [8.6]
0.10 0.12 [9] [9.3] 0.08 [3.3] 0.05 0.05 0.19 0.06 [11.2] [10.7] [19.6] [11.6] 0.03 [3.9] 0.21 0.22 0.15 0.22 Column (22) & Column (26) 0.00 [1.44] -0.04 [-3.59] -0.08 [-0.2]
0.09 0.10 0.12 [8.6] [9.3] [8] 0.11 [4.3] 0.14 0.13 0.15 0.14 [12] [11.5] [13.2] [11.9] -0.06 [-2.9] 0.27 0.28 0.26 0.27
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
45 Page 45 of 60
ip t
Class 1: small capital stock
zit2
0.21 0.20 0.19 0.19 [19.3] [18.8] [20.1] [19.1] 0.08 0.05 [4] [2.6]
FV zit*FV Adj. R2 0.22 0.22 0.31 0.31 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.02 [1.8] Class 1 zit2 -Class 2 zit2 0.03 [0.75] Class 1 FV - Class 2 FV qit qit2
0.29 [7.8]
0.28 0.12 0.11 [5.4] [10.9] [8.5] 0.01 0.04 [0.3] [2.5]
qit*FV
(5) 0.21 [18.4]
(8)
0.20 [18] 0.08 [4] -0.02 -0.02 [-1.3] [-1.3]
0.20 [18.1]
0.24 0.24 [6.7] [4.8] 0.01 [0.2] 0.54 0.54 [15.6] [15.6]
0.20 [5.5]
0.38 -0.03 [14] [-2] 0.03 [1.6] 0.23 0.23 0.10 0.23 Column (6) & Column (10) 0.03 [2.67] 0.03 [1.86] -0.09 [-2.58]
0.38 0.35 [14] [5.8] 0.18 [3.8] 0.14 0.14 0.10 0.14 Column (6) & Column (10) 0.14 [2.35] -0.02 [-2.01] 0.38 [0.79]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.05 0.26 0.26 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.18 [1.87] Class 1 qit2 -Class 2 qit2 -0.03 [1.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.18 0.18 0.20 [18.8] [17.7] [16.2] 0.05 [2.8] 0.07 0.07 0.16 0.11 [3.7] [3.8] [8.4] [3.9] -0.10 [-2] 0.32 0.32 0.23 0.32
0.11 0.10 0.16 [9.7] [7.8] [11] 0.03 [1.8] 0.15 0.15 0.16 0.30 [7.9] [7.8] [8.4] [9.3] -0.27 [-5.7] 0.28 0.28 0.23 0.29
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: large capital stock
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.20 0.20 0.19 0.15 0.14 0.14 0.18 0.17 0.19 0.17 0.16 0.12 [14.1] [14.1] [12.7] [12.1] [11.4] [8.3] [15.2] [15] [14.7] [16.7] [16] [9.1] 0.08 0.05 0.08 0.04 [3.9] [2.7] [3.6] [2.1] 0.00 0.00 0.16 0.00 0.03 0.03 0.07 0.02 0.03 0.03 0.19 0.04 0.09 0.08 0.15 0.06 [1.1] [0.7] [32.8] [0.3] [5.2] [5.3] [17.3] [3.1] [5.8] [5.6] [19.6] [5.9] [7.4] [7.2] [13.2] [5.2] 0.01 0.01 0.02 0.09 [2.3] [1.1] [2.5] [1.4] 0.22 0.22 0.31 0.22 0.32 0.32 0.29 0.32 0.23 0.23 0.15 0.23 0.33 0.33 0.26 0.33 Column (14) & Column (18) Column (22) & Column (26) 0.05 [2.16] 0.01 [2.19] 0.03 [1.76] 0.04 [1.52] -0.03 [-6.95] -0.05 [-2.12]
M an
Financial Variables
cr
Table 2 (cont'd) -Response of investment to alternative fundamentals and financial variables (when firms are grouped by the size of their capital stocks)
0.15 [4.9]
0.13 -0.06 0.09 0.08 0.07 0.18 0.17 0.18 [3] [-1.6] [8.9] [7] [4.1] [5.1] [3.4] [4.7] 0.02 0.03 0.01 [0.7] [2.1] [0.5] 0.18 0.18 0.16 0.11 0.06 0.06 0.07 0.05 0.21 0.21 0.19 0.21 [32.9] [32.9] [32.8] [11.2] [15.3] [15.3] [17.3] [13.5] [18.7] [18.7] [19.6] [9.5] 0.07 0.01 0.01 [9.6] [1.5] [1.8] 0.35 0.34 0.31 0.37 0.33 0.33 0.29 0.33 0.17 0.17 0.15 0.17 Column (14) & Column (18) Column (22) & Column (26) 0.05 [1.85] 0.07 [1.88] -0.01 [-1.85] -0.02 [-2.39] 0.12 [6.98] 0.11 [1.71]
0.11 0.09 0.10 [9.5] [7.3] [7.3] 0.04 [2.4] 0.11 0.11 0.15 0.11 [10] [10] [13.2] [7] 0.00 [0.1] 0.29 0.29 0.26 0.29
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
46 Page 46 of 60
ip t
Class 1: low number of employees
ait2
0.26 0.26 0.17 0.17 [17.1] [17.1] [13.2] [13.2] 0.18 0.30 [3.4] [6.2]
FV ait*FV Adj. R2 0.22 0.22 0.24 0.25 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.09 [2.24] Class 1 ait2 - Class 2 ait2 -0.13 [-2.96] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 [10.9] [11.5] 0.10 [4.4]
0.11 [11]
0.13 [12] 0.15 [5.5]
rait*FV
(8)
0.24 0.24 0.23 [16.2] [16.3] [14.9] 0.15 [2.9] 0.05 0.05 0.36 0.05 [4.7] [4.7] [13.4] [4.7] 0.06 [2.5] 0.23 0.23 0.09 0.23 Column (6) & Column (10) 0.09 [2.19] -0.17 [-2.06] -0.16 [-5.34] 0.12 [10.5]
0.13 0.11 [11] [9.8] 0.09 [3.8] 0.02 0.02 0.36 0.01 [1.6] [1.4] [13.4] [0.9] 0.05 [2.4] 0.19 0.20 0.09 0.20 Column (6) & Column (10) 0.03 [2.19] -0.05 [-4.59] -0.16 [-5.34]
ce pt
FV
(5)
Class 1 (6) (7)
Adj. R2 0.18 0.19 0.22 0.23 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.00 [1.74] Class 1 rait2 -Class 2 rait2 -0.04 [-1.84] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.15 0.15 0.21 [12.2] [12.1] [10] 0.32 [6.6] 0.21 0.21 0.24 0.23 [10.9] [11.2] [12.3] [11.5] -0.26 [-3.4] 0.27 0.28 0.23 0.27
0.21 0.21 0.11 0.12 0.12 0.10 0.23 0.23 0.15 0.14 0.14 0.10 [13.5] [13.6] [5.1] [9.2] [9.2] [4.5] [15.1] [15.2] [8.5] [11.3] [11.2] [5.7] 0.16 0.29 0.17 0.29 [3.2] [6.1] [3.3] [6] 0.02 0.02 0.17 0.03 0.05 0.05 0.06 0.05 0.04 0.04 0.19 0.04 0.13 0.13 0.14 0.13 [7.8] [7.7] [33.6] [8.9] [13.5] [13.5] [16.9] [12.9] [9.8] [9.7] [19.5] [10.2] [12.8] [12.8] [14.1] [12.9] 0.03 0.01 0.06 0.07 [6.9] [1.1] [7.1] [1.3] 0.24 0.24 0.32 0.25 0.29 0.30 0.27 0.29 0.25 0.25 0.15 0.26 0.28 0.29 0.25 0.28 Column (14) & Column (18) Column (22) & Column (26) 0.10 [2.65] 0.09 [2.99] -0.13 [-2.01] -0.12 [-2.01] -0.03 [-2.39] -0.09 [-8.82]
0.08 0.09 0.13 [7.5] [8.5] [7.6] 0.14 [5.1] 0.18 0.18 0.24 0.20 [8.7] [8.5] [12.3] [9.4] -0.23 [-3.6] 0.24 0.25 0.23 0.24
0.07 0.08 0.10 0.05 0.06 0.11 0.09 0.10 0.12 0.09 0.10 0.12 [6.1] [6.6] [6.1] [4.4] [5.2] [6.7] [8.3] [8.7] [9.2] [8.9] [9.7] [7.9] 0.07 0.11 0.08 0.12 [3.2] [4.3] [3.3] [4.6] 0.03 0.03 0.17 0.04 0.05 0.05 0.06 0.06 0.05 0.05 0.19 0.06 0.13 0.12 0.14 0.13 [9.9] [9.4] [33.6] [9.2] [13.4] [13] [16.9] [14.2] [10.9] [10.5] [19.5] [11.5] [12.3] [11.9] [14.1] [12.5] 0.01 -0.03 0.03 -0.04 [2.5] [-5] [4.1] [-2.5] 0.21 0.21 0.32 0.21 0.27 0.27 0.27 0.27 0.22 0.22 0.15 0.22 0.26 0.27 0.25 0.26 Column (14) & Column (18) Column (22) & Column (26) 0.02 [2.65] -0.01 [-0.99] -0.04 [-4.2] -0.05 [-4.09] -0.02 [-2.39] -0.08 [-8.82]
ed
ait
FV=CF_K
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high number of employees
M an
Financial Variables
cr
Table 3 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the number of employees)
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t -statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
47 Page 47 of 60
ip t
Class 1: low number of employees
zit2
0.20 0.20 0.19 0.19 [19.2] [18.8] [20.3] [19.1] 0.07 0.07 [3.5] [3.5]
FV zit*FV Adj. R2 0.23 0.23 0.30 0.31 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.01 [1.76] Class 1 zit2 -Class 2 zit2 0.00 [0.21] Class 1 FV - Class 2 FV qit qit2
0.30 [8.1]
0.30 0.11 0.09 [5.7] [10.2] [7.8] 0.00 0.05 [0] [3]
qit*FV
(5)
(8)
0.21 0.20 0.20 [18.7] [18.3] [18.4] 0.07 [3.5] -0.02 -0.02 0.36 -0.03 [-1.7] [-1.6] [13.4] [-2.2] 0.03 [2.5] 0.23 0.23 0.09 8.00 Column (6) & Column (10) 0.03 [1.76] 0.00 [2.24] -0.13 [-4.72] 0.25 [7]
0.25 0.21 [5] [5.7] 0.00 [0] 0.49 0.49 0.36 0.28 [14.6] [14.6] [13.4] [4.6] 0.21 [4.2] 0.13 0.13 0.09 0.13 Column (6) & Column (10) 0.17 [1.92] -0.02 [-1.82] 0.16 [0.87]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.05 0.24 0.24 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.21 [3.87] Class 1 qit2 -Class 2 qit2 -0.05 [-1.73] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.18 0.17 0.18 [17.4] [16.4] [13.9] 0.07 [3.5] 0.11 0.11 0.24 0.13 [5.6] [5.7] [12.3] [4.6] -0.05 [-0.9] 0.31 0.31 0.23 0.31
0.09 0.08 0.09 [8.2] [6.8] [6] 0.02 [1.2] 0.33 0.33 0.24 0.34 [14.8] [14.5] [12.3] [12.2] -0.01 [-0.3] 0.30 0.30 0.23 0.30
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high number of employees
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.19 0.19 0.19 0.16 0.15 0.15 0.18 0.17 0.19 0.17 0.17 0.14 [14.2] [14.1] [13.1] [12.9] [12.4] [9.2] [15.4] [15.1] [14.7] [16.7] [16] [10.4] 0.07 0.06 0.07 0.06 [3.4] [3.2] [3.2] [2.9] 0.00 0.00 0.17 0.00 0.02 0.02 0.06 0.02 0.03 0.03 0.19 0.04 0.08 0.08 0.14 0.07 [1.2] [0.9] [33.6] [0.5] [4.6] [4.4] [16.9] [2.2] [5.7] [5.5] [19.5] [5.6] [8] [7.7] [14.1] [5.5] 0.01 0.01 0.02 0.06 [2.1] [1] [2.2] [1.2] 0.23 0.23 0.32 0.22 0.31 0.31 0.27 0.31 0.23 0.24 0.15 0.24 0.32 0.32 0.25 0.32 Column (14) & Column (18) Column (22) & Column (26) 0.04 [1.74] 0.01 [1.94] 0.01 [1.62] 0.01 [1.3] -0.02 [-3.01] -0.05 [-7.31]
M an
Financial Variables
cr
Table 3 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the number of employees)
0.15 [4.8]
0.13 -0.04 0.09 0.08 0.06 0.19 0.18 0.19 [3] [-1.2] [8.6] [6.6] [3.7] [5.3] [3.6] [4.9] 0.02 0.04 0.01 [0.6] [2.5] [0.3] 0.19 0.19 0.17 0.11 0.05 0.05 0.06 0.05 0.21 0.21 0.19 0.21 [33.2] [33.2] [33.6] [11] [16.3] [16.2] [16.9] [14.6] [18.4] [18.4] [19.5] [9.2] 0.07 0.01 0.00 [8.9] [1.2] [2.1] 0.35 0.35 0.32 0.37 0.31 0.31 0.27 0.31 0.17 0.17 0.15 0.17 Column (14) & Column (18) Column (22) & Column (26) 0.06 [2.21] 0.10 [1.76] -0.02 [-1.61] -0.03 [-2.01] 0.13 [1.48] 0.09 [0.21]
0.09 0.08 0.09 [8.6] [6.5] [6.3] 0.04 [2.7] 0.13 0.13 0.14 0.12 [12] [11.9] [14.1] [9.9] 0.01 [0.3] 0.28 0.28 0.25 0.28
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
48 Page 48 of 60
ip t
Class 1: low dividend payout ratio
ait2
0.24 0.24 0.19 0.19 [15.3] [15.4] [18.2] [18.1] 0.17 0.32 [3.1] [7.6]
FV ait*FV Adj. R2 0.22 0.22 0.25 0.26 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.06 [6.29] Class 1 ait2 - Class 2 ait2 -0.15 [-3.01] Class 1 FV - Class 2 FV rait rait2
0.12 [10.4]
0.13 0.13 0.13 [11] [14.8] [15.6] 0.11 0.12 [4.3] [5.7]
rait*FV
(5)
(8)
Adj. R2 0.19 0.19 0.22 0.23 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.00 [4.4] Class 1 rait2 -Class 2 rait2 -0.01 [-1.66] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.23 0.23 0.22 [14.3] [14.4] [13.2] 0.14 [2.7] 0.07 0.07 0.39 0.06 [5.2] [5.2] [13.9] [4.4] 0.06 [2.6] 0.23 0.23 0.10 0.23 Column (6) & Column (10) 0.05 [4.03] -0.17 [-2.58] -0.05 [-2.81]
0.18 0.18 0.15 [17.6] [17.5] [12.7] 0.32 [7.7] 0.12 0.12 0.14 0.16 [9.4] [9.4] [10.5] [10.5] 0.09 [1.7] 0.27 0.29 0.18 0.27
0.19 0.19 0.08 0.17 0.17 0.09 0.20 0.20 0.12 0.19 0.19 0.17 [10.8] [10.9] [3] [16] [15.9] [5.7] [12.7] [12.8] [6] [17.8] [17.7] [13.3] 0.16 0.30 0.16 0.31 [2.9] [7.4] [3] [7.4] 0.03 0.03 0.19 0.02 0.03 0.03 0.03 0.03 0.06 0.06 0.23 0.05 0.02 0.02 0.03 0.03 [8.6] [8.5] [36.6] [6.5] [11.5] [11.3] [14.5] [13.6] [11] [10.9] [21.4] [9.2] [5.3] [5.1] [5.9] [6] 0.04 0.02 0.10 0.02 [5.7] [1.2] [6.9] [1.3] 0.24 0.24 0.36 0.25 0.28 0.30 0.21 0.29 0.25 0.26 0.18 0.27 0.25 0.27 0.16 0.25 Column (14) & Column (18) Column (22) & Column (26) 0.02 [2.13] 0.02 [1.83] -0.15 [-2.33] -0.15 [-2.46] 0.01 [1.95] 0.03 [4.89]
0.12 0.13 0.11 [9.7] [10.2] [8.9] 0.09 [3.7] 0.04 0.03 0.39 0.04 [2.4] [2.2] [13.9] [2.3] 0.03 [2.4] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.01 [1.69] -0.02 [-1.75] -0.05 [-5.04]
0.10 0.11 0.11 [11.4] [12.2] [10.9] 0.12 [5.5] 0.08 0.08 0.14 0.09 [5.7] [5.5] [10.5] [4.2] -0.01 [-0.3] 0.23 0.24 0.18 0.23
0.06 0.07 0.09 [4.8] [5.3] [4.9] 0.07 [2.6] 0.04 0.04 0.19 0.05 [11.1] [10.5] [36.6] [10.4] 0.01 [2.1] 0.22 0.23 0.36 0.22 Column (14) & Column (18) -0.03 [-0.77] -0.04 [-1.77] 0.02 [6.22]
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend payout ratio
M an
Financial Variables
cr
Table 4 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend payout ratio)
0.09 0.10 0.13 [10] [10.8] [10.3] 0.11 [5.3] 0.02 0.02 0.03 0.03 [9.6] [9.3] [14.5] [10.1] -0.02 [-4.6] 0.24 0.25 0.21 0.25
0.09 [7.6]
0.10 0.12 0.12 0.13 0.14 [8] [8.6] [13.7] [14.5] [12.2] 0.07 0.12 [2.8] [5.7] 0.07 0.07 0.23 0.08 0.02 0.02 0.03 0.03 [12.3] [11.9] [21.4] [12.9] [3.3] [3.2] [5.9] [4.1] 0.03 -0.02 [3.9] [-2.4] 0.23 0.23 0.18 0.24 0.22 0.23 0.16 0.22 Column (22) & Column (26) -0.03 [-0.61] -0.05 [-1.96] 0.05 [5.91]
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
49 Page 49 of 60
ip t
Class 1: low dividend payout ratio
zit2
0.22 0.21 0.17 0.17 [19.6] [18.9] [20.6] [20.4] 0.05 0.14 [2.3] [6.9]
FV zit*FV Adj. R2 0.25 0.25 0.28 0.29 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.04 [4.81] Class 1 zit2 -Class 2 zit2 -0.09 [-2.6] Class 1 FV - Class 2 FV qit qit2
0.33 [8.7]
0.35 0.07 0.06 [6.5] [7.6] [5.2] -0.01 0.04 [-0.4] [3.4]
qit*FV
(5)
(8)
Adj. R2 0.06 0.06 0.17 0.17 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.29 [3.87] Class 1 qit2 -Class 2 qit2 -0.05 [-2.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
0.16 [18]
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.21 0.21 0.21 [18.8] [18.2] [18.3] 0.05 [2.3] -0.01 -0.01 0.39 -0.01 [-0.4] [-0.4] [13.9] [-0.8] 0.02 [1.8] 0.25 0.25 0.10 0.25 Column (6) & Column (10) 0.05 [2.03] -0.09 [-2.74] -0.03 [-4.69]
0.16 0.16 [18] [17.4] 0.14 [6.8] 0.03 0.02 0.14 0.02 [1.9] [1.5] [10.5] [0.7] 0.02 [0.7] 0.28 0.29 0.18 0.28
0.20 0.20 0.21 0.15 0.15 0.15 0.18 0.18 0.20 0.17 0.17 0.16 [13.7] [13.4] [12.1] [15.4] [15.6] [13.2] [14.9] [14.5] [14.5] [19.5] [19.4] [16.2] 0.05 0.13 0.04 0.14 [2.2] [6.6] [2] [7] 0.01 0.01 0.19 0.01 0.01 0.01 0.03 0.00 0.05 0.05 0.23 0.06 0.00 0.00 0.03 -0.01 [1.2] [1.1] [36.6] [0.9] [3.4] [2.7] [14.5] [0.6] [7.4] [7.4] [21.4] [7.5] [-0.2] [-0.7] [5.9] [-2] 0.00 0.01 0.03 -0.02 [1.2] [1.2] [2.8] [-2.4] 0.25 0.25 0.36 0.24 0.28 0.29 0.21 0.28 0.26 0.26 0.18 0.26 0.28 0.29 0.16 0.28 Column (14) & Column (18) Column (22) & Column (26) 0.05 [2.64] 0.01 [2.05] -0.09 [-2.54] -0.10 [-2.23] 0.00 [2.75] 0.05 [6.73]
0.27 0.28 [7.4] [5.5] -0.01 [-0.3] 0.59 0.59 0.39 [16.1] [16.1] [13.9]
0.07 0.05 0.09 [6.9] [4.9] [8.7] 0.03 [2.8] 0.13 0.13 0.14 0.25 [9.2] [9] [10.5] [10.4] -0.10 [-6.2] 0.20 0.20 0.18 0.21
0.15 0.12 -0.35 0.06 0.05 0.11 0.20 0.18 0.16 [4.8] [2.8] [-10.1] [6.8] [4.8] [9.4] [5.6] [3.6] [4.2] 0.02 0.03 0.02 [0.9] [2.7] [0.6] 0.22 0.22 0.19 0.05 0.03 0.03 0.03 0.05 0.27 0.27 0.23 0.21 [37.3] [37.3] [36.6] [5.6] [12.9] [12.7] [14.5] [13.6] [21.2] [21.2] [21.4] [8.6] 0.19 -0.02 0.05 [23.7] [-6.6] [2.6] 0.41 0.41 0.36 0.52 0.22 0.22 0.21 0.24 0.21 0.21 0.18 0.21 Column (14) & Column (18) Column (22) & Column (26) 0.07 [2.48] 0.13 [2.78] -0.01 [-1.97] -0.02 [-3.15] 0.19 [13.77] 0.25 [7.2]
0.19 [5]
0.18 [3] 0.46 [8] 0.15 0.15 0.10 0.17 Column (6) & Column (10) 0.23 [3.3] -0.04 [-2.57] 0.46 [10.44]
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend payout ratio
M an
Financial Variables
cr
Table 4 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend payout ratio)
0.07 [7.1]
0.02 [4.6]
0.18
0.06 0.07 [5] [6.8] 0.04 [2.9] 0.02 0.03 0.03 [4.3] [5.9] [2.9] 0.00 [-0.5] 0.18 0.16 0.18
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
50 Page 50 of 60
ip t
Class 1: low dividend to capital ratio
ait2
0.23 0.23 0.21 0.21 [14.6] [14.6] [19.2] [19.6] 0.13 0.41 [2.5] [9]
FV ait*FV Adj. R2 0.21 0.21 0.27 0.29 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.01 [5.07] Class 1 ait2 - Class 2 ait2 -0.28 [-2.01] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 0.13 0.14 [10.2] [10.8] [14.6] [15.5] 0.12 0.12 [4.6] [5.3]
rait*FV
(5)
(8)
Adj. R2 0.19 0.19 0.23 0.24 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait -0.01 [-1.52] Class 1 rait2 -Class 2 rait2 0.00 [0.7] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.23 0.23 0.21 [13.9] [13.9] [13.1] 0.23 [2.1] 0.06 0.06 0.41 0.05 [4.2] [4.3] [14.1] [3.7] 0.04 [2.5] 0.22 0.22 0.10 0.22 Column (6) & Column (10) 0.06 [2.79] 0.08 [4.51] -0.08 [-7.75]
0.17 0.17 0.15 [17.8] [18.2] [12] 0.15 [8.6] 0.15 0.14 0.16 0.17 [12.4] [12.1] [13.1] [13.7] 0.11 [1.2] 0.31 0.33 0.23 0.32
0.12 0.13 0.12 [10.1] [10.6] [9.3] 0.10 [4] 0.02 0.02 0.41 0.02 [1.4] [1.1] [14.1] [1.2] 0.03 [1.3] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.03 [0.61] -0.01 [-0.7] -0.10 [-2.92]
0.09 0.10 0.10 [9.5] [10.4] [9.6] 0.11 [4.9] 0.12 0.12 0.16 0.14 [9] [8.8] [13.1] [8.1] -0.06 [-1.9] 0.25 0.26 0.23 0.25
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend capital ratio
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.19 0.20 0.07 0.15 0.15 0.09 0.20 0.20 0.10 0.17 0.17 0.17 [10.5] [10.6] [2.8] [16] [16.4] [5.6] [12.3] [12.3] [5.2] [18.3] [18.6] [12.8] 0.25 0.13 0.25 0.13 [2.4] [8.5] [2.6] [8.4] 0.03 0.03 0.21 0.03 0.03 0.03 0.04 0.03 0.06 0.06 0.26 0.05 0.04 0.04 0.05 0.04 [8.1] [8.1] [38.2] [6.6] [13.4] [13.1] [16.6] [15.5] [10.1] [10.1] [22.7] [9.6] [10.8] [10.4] [12.2] [11.5] 0.04 0.02 0.11 0.03 [5.7] [0.9] [7.3] [0.5] 0.23 0.23 0.38 0.24 0.32 0.34 0.25 0.33 0.24 0.24 0.20 0.26 0.30 0.32 0.22 0.31 Column (14) & Column (18) Column (22) & Column (26) 0.05 [2.77] 0.03 [2.45] 0.11 [4.12] 0.12 [4.07] 0.00 [4.59] 0.02 [1.63]
M an
Financial Variables
cr
Table 5 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend to capital ratio)
0.07 0.07 0.09 [5.2] [5.7] [5.2] 0.07 [3] 0.04 0.04 0.21 0.05 [10.4] [9.7] [38.2] [10] 0.01 [2.2] 0.22 0.22 0.38 0.22 Column (14) & Column (18) -0.02 [-0.99] -0.03 [-1.54] 0.01 [4.57]
0.08 0.09 0.13 [8.3] [9.1] [9.3] 0.10 [4.7] 0.03 0.03 0.04 0.04 [12] [11.8] [16.6] [12] -0.02 [-4.9] 0.27 0.28 0.25 0.28
0.09 [7.8]
0.10 0.12 0.11 0.12 0.14 [8.2] [8.7] [12.3] [13.1] [11.8] 0.08 0.11 [3.1] [5] 0.07 0.07 0.26 0.08 0.04 0.04 0.05 0.05 [11.4] [10.9] [22.7] [12.1] [9.3] [9.1] [12.2] [9.6] 0.04 -0.03 [4] [-3.7] 0.23 0.23 0.20 0.23 0.25 0.26 0.22 0.26 Column (22) & Column (26) -0.03 [-0.73] -0.03 [-1.21] 0.03 [2.53]
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
51 Page 51 of 60
ip t
Class 1: low dividend to capital ratio
zit2
0.22 0.21 0.17 0.16 [19.9] [19.1] [19.8] [19.4] 0.06 0.09 [3] [4.6]
FV zit*FV Adj. R2 0.25 0.25 0.28 0.28 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.05 [2.43] Class 1 zit2 -Class 2 zit2 -0.02 [-1.78] Class 1 FV - Class 2 FV qit qit2
0.33 [8.5]
0.33 0.08 0.07 [6.1] [8.4] [6.2] 0.00 0.03 [0] [2.4]
qit*FV
(5)
(8)
Adj. R2 0.06 0.06 0.23 0.23 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.26 [1.87] Class 1 qit2 -Class 2 qit2 -0.03 [-2.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.21 0.21 0.21 [19.2] [18.5] [18.7] 0.06 [3] -0.01 0.00 0.41 -0.01 [-0.4] [-0.3] [14.1] [-0.9] 0.02 [2.1] 0.26 0.26 0.10 0.26 Column (6) & Column (10) 0.05 [1.71] -0.02 [-2.4] -0.03 [-3.81]
0.16 0.16 0.16 [16.7] [16.4] [16] 0.08 [4.5] 0.03 0.03 0.16 0.03 [2.2] [2] [13.1] [1.4] 0.00 [0.2] 0.28 0.28 0.23 0.28
0.27 0.28 0.16 [7.4] [5.5] [4.4] -0.01 [-0.2] 0.62 0.62 0.41 0.10 [16.7] [16.7] [14.1] [1.6] 0.61 [10] 0.16 0.16 0.10 0.19 Column (6) & Column (10) 0.22 [3] -0.03 [-3.13] 0.50 [10.93]
0.07 0.06 0.09 [7.3] [5.5] [8.7] 0.02 [1.9] 0.13 0.13 0.16 0.23 [9.5] [9.4] [13.1] [9.7] -0.08 [-5.3] 0.22 0.22 0.23 0.23
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: high dividend capital ratio
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.20 0.20 0.21 0.15 0.14 0.14 0.18 0.18 0.20 0.17 0.16 0.15 [13.9] [13.7] [12.5] [14.1] [13.9] [11.8] [15.3] [14.8] [15] [18.3] [18] [15] 0.06 0.08 0.06 0.09 [2.9] [4.4] [2.7] [4.6] 0.01 0.01 0.21 0.01 0.01 0.01 0.04 0.00 0.05 0.05 0.26 0.07 0.00 0.00 0.05 -0.01 [1.5] [1.2] [38.2] [1.4] [3.6] [3.4] [16.6] [0.2] [8] [7.9] [22.7] [8.1] [0.6] [0.2] [12.2] [-1.5] 0.00 0.01 0.03 0.02 [1.6] [0.8] [3.1] [1.4] 0.25 0.25 0.38 0.25 0.28 0.28 0.25 0.28 0.27 0.27 0.20 0.27 0.27 0.28 0.22 0.28 Column (14) & Column (18) Column (22) & Column (26) 0.06 [2.04] 0.01 [1.69] -0.02 [-2] -0.03 [-2.21] 0.00 [2.08] 0.05 [10.4]
M an
Financial Variables
cr
Table 5 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the dividend to capital ratio)
0.14 0.11 -0.37 0.07 0.06 0.11 0.21 0.19 0.16 [4.6] [2.6] [-10.8] [7.4] [5.6] [9.2] [5.9] [3.9] [4.3] 0.03 0.02 0.01 [1.1] [1.8] [0.5] 0.23 0.23 0.21 0.05 0.03 0.03 0.04 0.05 0.29 0.29 0.26 0.22 [38.8] [38.8] [38.2] [6.1] [13.7] [13.6] [16.6] [13.2] [22.3] [22.3] [22.7] [8.7] 0.20 -0.01 0.06 [24.3] [-5.4] [3] 0.42 0.42 0.38 0.54 0.25 0.25 0.25 0.26 0.22 0.22 0.20 0.22 Column (14) & Column (18) Column (22) & Column (26) 0.05 [0.68] 0.13 [2.21] 0.01 [2.13] -0.01 [-3.25] 0.20 [13.62] 0.26 [6.68]
0.08 0.06 0.08 [7.4] [5.5] [7.1] 0.02 [2] 0.03 0.03 0.05 0.03 [5.8] [5.6] [12.2] [3.5] 0.00 [-0.4] 0.20 0.21 0.22 0.20
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
52 Page 52 of 60
ip t
Class 1: firms without debt rating
ait2
0.25 0.25 0.25 0.13 [20.7] [20.8] [16.1] [7.6] 0.20 0.27 [4.8] [3.9]
FV ait*FV Adj. R2 0.23 0.23 0.19 0.24 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.12 [2.29] Class 1 ait2 - Class 2 ait2 -0.07 [-2.4] Class 1 FV - Class 2 FV rait rait2
0.12 0.13 0.13 0.11 [13.6] [14.4] [10.8] [7.8] 0.11 0.18 [5.5] [4.9]
rait*FV
Adj. R2 0.19 0.19 0.15 0.23 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.02 [2.55] Class 1 rait2 -Class 2 rait2 -0.07 [-2.06] Class 1 FV - Class 2 FV
FV=Sales_K
Class 1 (13) (14) (15) (16)
0.23 0.23 0.22 [19.7] [19.8] [17.9] 0.18 [4.4] 0.07 0.07 0.34 0.06 [6.5] [6.5] [15.7] [6.5] 0.05 [2.4] 0.24 0.24 0.10 0.24 Column (6) & Column (10) 0.14 [1.79] -0.06 [-4.84] -0.40 [-6.76]
0.23 0.09 0.18 [15.5] [5.7] [5.7] 0.24 [3.6] 0.06 0.46 0.52 0.48 [4.2] [11.3] [12.8] [11.7] -0.41 [-3.3] 0.20 0.30 0.28 0.30
0.20 0.20 0.10 [16] [16.1] [5.5] 0.19 [4.5] 0.03 0.03 0.16 0.03 [10.7] [10.5] [38.7] [11.7] 0.03 [8.2] 0.25 0.25 0.31 0.26 Column (14) & Column (18) 0.11 [2.28] -0.08 [-4.47] -0.02 [-2.39]
0.22 0.22 0.15 0.20 [18.4] [18.4] [10.5] [12.4] 0.19 [4.7] 0.05 0.05 0.19 0.05 0.10 [13.2] [13.1] [23.7] [13.7] [11] 0.06 [7.7] 0.26 0.27 0.16 0.27 0.22 Column (22) & Column (26) 0.10 [2.45] -0.07 [-4.39] 0.03 [6.92]
0.12 [6.9] 0.27 [3.8] 0.02 [1.1]
0.12 0.13 0.11 [12.6] [13.3] [11.6] 0.09 [4.8] 0.04 0.03 0.34 0.03 [3] [2.7] [15.7] [2.3] 0.05 [2.4] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.09 [2.64] -0.06 [-4.42] -0.42 [-7.54]
0.14 0.04 0.09 [11.2] [2.5] [4] 0.15 [4.1] 0.00 0.46 0.52 0.50 [0.2] [9.6] [12.8] [10.5] -0.35 [-3.8] 0.17 0.28 0.28 0.28
0.07 0.08 0.11 0.07 0.06 0.09 0.09 0.10 0.13 0.09 [7.2] [7.9] [7.8] [5.3] [3.9] [4.1] [10.5] [11.1] [12.1] [7.5] 0.08 0.15 0.08 [3.9] [4.2] [4] 0.03 0.03 0.16 0.04 0.04 0.05 0.06 0.06 0.06 0.06 0.19 0.07 0.13 [12.8] [12.2] [38.7] [12.5] [10.7] [8.5] [11.3] [9.3] [14.2] [13.7] [23.7] [15.5] [13.5] 0.02 -0.02 0.04 [3.9] [-2.8] [2.2] 0.22 0.22 0.31 0.22 0.19 0.27 0.27 0.26 0.23 0.23 0.16 0.24 0.21 Column (14) & Column (18) Column (22) & Column (26) 0.02 [1.7] -0.01 [-1.01] -0.08 [-3.56] -0.12 [-2.23] -0.02 [-2.91] 0.03 [4.62]
0.11 [7.3] 0.19 [5] 0.02 [1.2]
(5)
(8)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K
Class 2 (9) (10) (11) (12)
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
Class 2: firms with debt rating
us
FV=None Class 1 (1) (2)
M an
Financial Variables
cr
Table 6 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by bond ratings)
0.19 0.09 0.16 [12] [5.4] [5.6] 0.27 [4.1] 0.03 0.05 0.06 0.05 [8.6] [9.2] [11.3] [9.7] -0.03 [-3.1] 0.21 0.28 0.27 0.27
Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.24
0.23
0.07 [3]
0.05 0.01 [3] [0.5] 0.12 [1.1] 0.21 0.23
0.06 [2.9]
0.05 0.02 [3] [1.2] 0.07 [1.4] 0.21 0.22
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
53 Page 53 of 60
ip t
Class 1: firms without debt rating
zit2
0.20 0.20 0.23 0.18 [23.7] [22.9] [20.3] [13.7] 0.07 0.08 [3.9] [3]
FV zit*FV Adj. R2 0.24 0.24 0.22 0.29 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.02 [2.36] Class 1 zit2 -Class 2 zit2 -0.01 [-1.03] Class 1 FV - Class 2 FV qit qit2
0.25 [9.2]
0.22 0.39 0.10 [6.1] [8.9] [5.8] 0.03 0.03 [1.3] [1.6]
qit*FV
(5)
(8)
0.20 [22.7]
0.20 0.20 [22] [22.2] 0.07 [4] -0.01 -0.01 0.34 -0.02 [-0.9] [-0.8] [15.7] [-1.5] 0.02 [1.4] 0.25 0.25 0.10 0.25 Column (6) & Column (10) 0.06 [2.44] -0.01 [-0.99] -0.32 [-8.67] 0.20 [7.7]
0.18 0.16 [5] [5.9] 0.03 [1.1] 0.48 0.48 0.34 0.28 [17.4] [17.4] [15.7] [5.9] 0.21 [5.2] 0.13 0.26 0.10 0.13 Column (6) & Column (10) 0.09 [1.14] 0.02 [2.31] 0.05 [1.54]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.23 0.05 0.05 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.13 [2.87] Class 1 qit2 -Class 2 qit2 0.00 [1.73] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.22 0.14 0.08 [19.7] [9.7] [3.7] 0.07 [2.8] 0.00 0.31 0.52 0.27 [0.1] [6.8] [12.8] [5.7] 0.29 [1.4] 0.23 0.31 0.28 0.32
0.19 0.19 0.19 0.21 0.15 0.16 0.17 0.17 0.19 0.19 0.19 [17] [16.7] [15.1] [13.7] [9.2] [7.2] [18.8] [18.3] [18] [15.4] [13.1] 0.07 0.08 0.06 0.09 [3.8] [2.9] [3.6] [3.1] 0.01 0.01 0.16 0.01 0.01 0.02 0.06 0.02 0.04 0.04 0.19 0.05 0.08 -0.03 [2.2] [1.9] [38.7] [1] [1.9] [3] [11.3] [2.1] [8.2] [8] [23.7] [7.9] [8.9] [-1.5] 0.00 0.00 0.02 [0.2] [0.1] [3] 0.24 0.24 0.31 0.24 0.22 0.30 0.27 0.29 0.25 0.26 0.16 0.25 0.25 0.29 Column (14) & Column (18) Column (22) & Column (26) 0.03 [2.32] -0.02 [-0.57] -0.01 [-1.74] -0.03 [-1.59] -0.01 [-0.24] 0.06 [1.7]
0.33 0.09 0.08 [8.1] [5.8] [3.8] 0.01 [0.4] 0.76 0.43 0.52 0.41 [18.6] [12.2] [12.8] [9] 0.05 [0.6] 0.17 0.17 0.28 0.33
0.12 0.09 -0.08 0.18 0.09 0.06 0.16 0.13 0.15 0.24 [5.3] [2.9] [-2.9] [5.2] [5.5] [2.9] [6] [3.7] [5.4] [6.4] 0.04 0.02 0.03 [1.7] [1.3] [1.3] 0.18 0.18 0.16 0.10 0.23 0.05 0.06 0.04 0.21 0.21 0.19 0.21 0.56 [38.7] [38.7] [38.7] [13.7] [40.1] [10.4] [11.3] [8.7] [22.4] [22.4] [23.7] [11.3] [33.6] 0.07 0.01 0.01 [12.2] [1.2] [0.5] 0.34 0.28 0.31 0.37 0.43 0.43 0.27 0.32 0.17 0.25 0.16 0.17 0.36 Column (14) & Column (18) Column (22) & Column (26) 0.00 [1.5] 0.03 [0.23] 0.01 [1.84] 0.00 [2.08] 0.13 [4.68] 0.13 [0.41]
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: firms with debt rating
M an
Financial Variables
cr
Table 6 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by bond ratings)
0.10 [5.7] 0.03 [1.3] 0.08 [5.5]
0.36
0.16 [9]
0.05 -0.04 [3] [-2] 0.06 [1.5] 0.21 0.29
0.08 [4]
0.05 0.04 [3] [1.5] 0.08 [0.8] 0.21 0.28
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
54 Page 54 of 60
ip t
Class 1: high debt ratio
ait2
0.25 0.25 0.19 0.19 [16.1] [16.2] [15.1] [15.1] 0.26 0.15 [4.8] [3.3]
FV ait*FV Adj. R2 0.19 0.19 0.28 0.28 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.06 [5.07] Class 1 ait2 - Class 2 ait2 0.11 [2.01] Class 1 FV - Class 2 FV rait rait2
0.13 0.14 0.11 0.11 [10.8] [11.6] [11.7] [12.3] 0.15 0.08 [5.7] [4]
rait*FV
(5)
(8)
Adj. R2 0.15 0.16 0.25 0.26 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.02 [2.52] Class 1 rait2 -Class 2 rait2 0.07 [2.7] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.23 0.23 0.13 [15.5] [15.6] [9.7] 0.23 [4.3] 0.06 0.06 0.14 0.12 [4.2] [4.2] [12.5] [10.5] 0.12 [5.6] 0.20 0.20 0.25 0.31 Column (6) & Column (10) 0.06 [2.79] 0.08 [4.51] -0.05 [-7.75]
0.17 0.17 0.24 [13.7] [13.7] [15.1] 0.15 [3.2] 0.11 0.11 0.51 0.06 [9.7] [9.7] [15.5] [4.3] -0.03 [-0.9] 0.30 0.31 0.11 0.20
0.14 0.15 0.08 [11.2] [11.8] [8.1] 0.13 [5] 0.00 0.00 0.14 0.12 [0.2] [0.2] [12.5] [7.5] 0.05 [2] 0.17 0.17 0.25 0.27 Column (6) & Column (10) 0.06 [1.71] 0.07 [0.7] -0.09 [-2.92]
0.08 0.08 0.13 [7.9] [8.4] [10] 0.07 [3.1] 0.10 0.10 0.51 0.01 [8.1] [7.7] [15.5] [0.5] 0.05 [1.3] 0.27 0.27 0.11 0.17
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: low debt ratio
Class 1 (13) (14) (15) (16) 0.19 0.20 [12] [12.1] 0.25 [4.6] 0.03 0.03 [8.6] [8.6]
0.06 [3.7]
0.07 0.08 [5.3] [6.1] 0.12 [4.4] 0.04 0.04 [10.7] [10.1]
0.14 [10.2]
M an
Financial Variables
cr
Table 7 - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the debt to capital ratio)
0.04 0.03 [16] [13.3] 0.03 [7.3] 0.21 0.21 0.28 0.33 Column (14) & Column (18) 0.05 [2.77] 0.11 [4.12] 0.00 [4.59]
0.04 0.05 [16] [13.6] 0.04 [8] 0.19 0.19 0.28 0.30 Column (14) & Column (18) 0.01 [1.99] 0.06 [1.54] 0.01 [4.57]
FV=WorkingK_K
Class 2 (17) (18) (19) (20)
Class 1 (21) (22) (23) (24)
0.15 0.15 0.09 [11.8] [11.8] [3.7] 0.13 [3] 0.03 0.03 0.22 0.03 [11.5] [11.4] [40.8] [7.1] 0.04 [1.2] 0.31 0.31 0.40 0.22
0.20 0.20 0.10 0.17 0.17 0.14 [12.4] [12.5] [6.8] [14] [14] [6.7] 0.25 0.13 [4.7] [3] 0.10 0.10 0.05 0.04 0.04 0.04 0.49 0.09 [11] [11] [14.7] [13.3] [12.3] [12.2] [32.7] [9.9] 0.06 0.08 [9.6] [1.6] 0.22 0.23 0.27 0.34 0.32 0.32 0.31 0.23 Column (22) & Column (26) 0.03 [2.45] 0.12 [4.07] 0.06 [1.63]
0.06 0.07 0.07 [6.5] [6.9] [3.8] 0.06 [2.7] 0.03 0.03 0.22 0.04 [11.4] [11] [40.8] [9.8] 0.00 [-0.4] 0.29 0.29 0.40 0.19
Class 2 (25) (26) (27) (28)
0.09 [7.5]
0.10 0.11 0.08 0.09 0.11 [8.3] [9.6] [9.1] [9.5] [6.6] 0.12 0.05 [4.5] [2.6] 0.13 0.12 0.05 0.05 0.04 0.04 0.49 0.13 [13.5] [13] [14.7] [11.8] [11.7] [11.3] [32.7] [13.6] 0.02 -0.03 [3.6] [-1.8] 0.21 0.21 0.27 0.29 0.29 0.29 0.31 0.21 Column (22) & Column (26) 0.01 [1.73] 0.07 [1.21] 0.08 [2.53]
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
55 Page 55 of 60
ip t
Class 1: high debt ratio
zit2
0.23 0.22 [20.3] [19.5] 0.05 [2.5]
0.17 0.16 [19] [18.6] 0.09 [4.9]
FV zit*FV Adj. R2 0.22 0.23 0.30 0.30 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.06 [2.43] Class 1 zit2 -Class 2 zit2 -0.04 [-1.78] Class 1 FV - Class 2 FV qit qit2
0.39 [8.9]
0.32 0.10 0.09 [5.4] [9.6] [6.7] 0.08 0.02 [1.7] [1.8]
qit*FV
(5)
(8)
0.22 0.22 0.16 [19.7] [18.9] [15.8] 0.05 [2.5] 0.00 0.00 0.14 -0.03 [0.1] [0.2] [12.5] [-1.4] 0.06 [2.5] 0.23 0.23 0.25 0.30 Column (6) & Column (10) 0.06 [1.71] -0.04 [-2.4] -0.01 [-3.81] 0.33 [8.1]
0.28 0.10 [5] [9.5] 0.06 [1.5] 0.76 0.76 0.14 0.23 [18.6] [18.5] [12.5] [9.5] 0.07 [4.6] 0.17 0.17 0.25 0.26 Column (6) & Column (10) 0.20 [3] 0.05 [3.13] 0.62 [10.93]
ce pt
FV
FV=CF_K Class 1 (6) (7)
Adj. R2 0.05 0.05 0.23 0.23 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.23 [0.87] Class 1 qit2 -Class 2 qit2 0.06 [1.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12) 0.16 0.16 0.22 [16.8] [16.5] [19.2] 0.09 [4.9] 0.01 0.01 0.51 0.00 [0.9] [0.7] [15.5] [-0.2] 0.02 [0.8] 0.30 0.30 0.11 0.23
0.09 [8.4]
0.08 0.11 [6] [2.6] 0.01 [1.5] 0.14 0.14 0.51 -0.36 [10.2] [10.2] [15.5] [-5.2] 1.50 [1.1] 0.26 0.26 0.11 0.28
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: low debt ratio
FV=WorkingK_K
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
Class 1 (21) (22) (23) (24)
0.21 0.21 [13.7] [13.5] 0.05 [2.3] 0.01 0.01 [1.9] [1.7]
0.15 [14.2]
0.19 0.18 0.14 0.15 0.15 0.20 [15.4] [14.9] [13.4] [15.6] [15.3] [13.5] 0.04 0.08 [1.9] [4.6] 0.08 0.08 0.05 0.01 0.02 0.02 0.49 0.09 [8.9] [8.7] [14.7] [2] [5.2] [4.9] [32.7] [8.8] 0.01 -0.03 [1.8] [-1.9] 0.25 0.25 0.27 0.30 0.30 0.31 0.31 0.25 Column (22) & Column (26) 0.04 [2.69] -0.04 [-1.21] 0.06 [10.4]
M an
Financial Variables
cr
Table 7 (cont'd) - Response of investment to alternative fundamentals and financial variables (when firms are grouped by the debt to capital ratio)
0.15 [12.8]
0.01 [1] 0.00 [0.2] 0.22 0.23 0.28 0.30 Column (14) & Column (18) 0.06 [2.04] -0.04 [-2] 0.00 [6.08] 0.18 0.10 [5.2] [2.1] 0.09 [2.6] 0.23 0.23 [40.1] [40.1]
0.04 [16]
0.13 [10.5]
0.04 0.06 [16] [12.6] 0.02 [6.1] 0.43 0.43 0.28 0.29 Column (14) & Column (18) 0.02 [1.88] 0.08 [2.13] 0.20 [13.62]
0.15 0.21 [14] [11.9] 0.09 [4.8] 0.01 0.01 0.22 0.01 [2.6] [2.4] [40.8] [1.5] 0.00 [-0.4] 0.30 0.30 0.40 0.22
0.09 0.08 -0.68 [8.4] [5.9] [-18.9] 0.01 [1.6] 0.04 0.04 0.22 0.01 [13] [12.9] [40.8] [1.9] 0.28 [0.9] 0.28 0.28 0.40 0.64
0.24 [6.4]
0.16 [3.2] 0.08 [2.2] 0.56 0.56 0.05 [33.6] [33.7] [14.7]
Class 2 (25) (26) (27) (28)
0.09 [8.3]
0.09 [8.2]
0.05 [6] 0.01 [1.3] 0.36 0.36 0.27 0.25 Column (22) & Column (26) 0.09 [2.21] 0.07 [3.25] 0.52 [6.68]
0.04 [9.4]
0.25
0.07 -0.37 [5.8] [-10.5] 0.01 [1.5] 0.04 0.49 -0.08 [9.4] [32.7] [-3.5] 0.83 [0.5] 0.25 0.31 0.57
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of their average capital stock.
56 Page 56 of 60
ip t
Class 1: Equity-dependent firms
ait2
0.24 0.24 0.12 0.12 [16.4] [16.5] [14.9] [14.9] 0.14 0.19 [2.8] [7.1]
FV ait*FV Adj. R2 0.19 0.19 0.25 0.27 Differences in coefficients Column (2) & Column (4) Class 1 ait - Class 2 ait 0.12 [2.25] Class 1 ait2 - Class 2 ait2 -0.05 [-1.71] Class 1 FV - Class 2 FV rait rait2
0.13 0.14 0.11 0.11 [11.8] [12.9] [9.6] [9.8] 0.14 0.06 [6.1] [2.3]
rait*FV
(5)
(8)
0.24 [15.9]
Adj. R2 0.17 0.18 0.22 0.22 Differences in coefficients Column (2) & Column (4) Class 1 rait - Class 2 rait 0.03 [2.91] Class 1 rait2 -Class 2 rait2 0.08 [1.77] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.24 0.24 [16] [15.5] 0.11 [2.4] 0.06 0.06 0.49 0.06 [4.4] [4.4] [15.4] [4.3] 0.01 [2.3] 0.20 0.20 0.10 0.20 Column (6) & Column (10) 0.13 [2.33] -0.07 [-2.78] 0.04 [4.9]
0.11 0.12 0.11 [14.5] [14.6] [13.7] 0.19 [7.1] 0.02 0.02 0.04 0.02 [1.9] [1.7] [3.6] [2] -0.03 [-1.2] 0.25 0.27 0.16 0.25
0.14 0.15 0.13 [12.7] [13.6] [11.5] 0.12 [5.5] -0.01 -0.01 0.49 -0.01 [-0.6] [-0.7] [15.4] [-0.3] 0.06 [2.2] 0.19 0.19 0.10 0.19 Column (6) & Column (10) 0.08 [2.58] 0.08 [0.51] -0.11 [-0.31]
0.07 0.07 0.08 [5.8] [5.9] [5.9] 0.04 [1.5] 0.10 0.10 0.04 0.12 [6.8] [6.6] [3.6] [6.5] -0.04 [-1.4] 0.23 0.23 0.16 0.23
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
ait
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: Not equity-dependent firms
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.21 0.22 0.18 0.14 0.12 0.04 0.23 0.24 0.18 0.11 0.11 0.11 [12.7] [12.8] [3.5] [10.7] [10.8] [1.8] [12.8] [12.8] [8.1] [13.5] [13.5] [6.5] 0.14 0.20 0.14 0.22 [2.8] [6.5] [2.9] [6.8] 0.03 0.03 0.23 0.02 0.04 0.03 0.04 0.04 0.11 0.11 0.51 0.10 0.04 0.04 0.05 0.05 [8] [7.9] [41.4] [6.4] [12.9] [12.5] [16.4] [14.7] [12.3] [12.4] [33.9] [11.6] [11.7] [11.5] [13.4] [12.5] 0.04 0.04 0.04 0.07 [6] [0.6] [1.9] [0.8] 0.21 0.21 0.41 0.22 0.30 0.31 0.24 0.32 0.23 0.24 0.32 0.24 0.29 0.30 0.21 0.31 Column (14) & Column (18) Column (22) & Column (26) 0.10 [3.35] 0.13 [3.23] -0.06 [-2.84] -0.08 [-3.31] 0.00 [3.52] 0.06 [10.99]
M an
Financial Variables
cr
Table 8 -Response of investment to alternative fundamentals and financial variables (when firms are grouped by their KZ index)
0.08 0.10 0.07 0.03 0.03 0.14 0.09 0.10 0.11 [7.3] [8.3] [3.9] [2.3] [2.4] [8.9] [8.6] [9.4] [7.7] 0.11 0.03 0.10 [4.8] [1] [4.3] 0.04 0.03 0.23 0.04 0.04 0.04 0.04 0.07 0.13 0.13 0.51 0.14 [9.4] [8.6] [41.4] [8.1] [15.2] [15.1] [16.4] [18.5] [14.7] [14] [33.9] [14.8] 0.01 -0.05 0.03 [0.8] [-10.5] [2.1] 0.20 0.20 0.41 0.20 0.28 0.28 0.24 0.31 0.23 0.24 0.32 0.24 Column (14) & Column (18) Column (22) & Column (26) 0.07 [2.53] 0.02 [1.82] 0.09 [1.97] 0.06 [1.74] -0.01 [-3.43] 0.09 [11.47]
0.08 0.08 0.10 [7] [7.1] [7.6] 0.04 [1.3] 0.04 0.04 0.05 0.05 [9.9] [9.7] [13.4] [10.1] -0.02 [-3.2] 0.25 0.25 0.21 0.25
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the ones taking place in the higher (lower) 50 percentile of the firms when they are sorted by the size of their KZ index.
57 Page 57 of 60
ip t
Class 1: Equity-dependent firms
zit2
0.21 0.21 0.19 0.18 [19.7] [19.2] [19.7] [19.1] 0.03 0.13 [1.4] [6.5]
FV zit*FV Adj. R2 0.22 0.22 0.29 0.30 Differences in coefficients Column (2) & Column (4) Class 1 zit - Class 2 zit 0.02 [2.26] Class 1 zit2 -Class 2 zit2 -0.10 [-2.66] Class 1 FV - Class 2 FV qit qit2
0.35 [8.7]
0.27 0.11 0.09 [5.5] [9.5] [5.7] 0.11 0.02 [2.6] [2.5]
qit*FV
(5)
(8)
Adj. R2 0.05 0.05 0.21 0.22 Differences in coefficients Column (2) & Column (4) Class 1 qit - Class 2 qit 0.19 [1.87] Class 1 qit2 -Class 2 qit2 0.08 [1.23] Class 1 FV - Class 2 FV
FV=Sales_K
Class 2 (9) (10) (11) (12)
0.20 0.20 0.20 [19.2] [18.6] [19.1] 0.03 [1.4] 0.00 0.00 0.49 0.00 [0.2] [0.3] [15.4] [0.3] 0.08 [3.3] 0.23 0.23 0.10 0.22 Column (6) & Column (10) 0.02 [2.61] -0.10 [-3.69] 0.01 [2.03]
0.19 0.18 0.18 [17.5] [17.2] [19.1] 0.13 [6.5] 0.00 0.00 0.04 -0.05 [0.2] [-0.1] [3.6] [-2.4] 0.00 [0.3] 0.29 0.30 0.16 0.29
0.28 0.22 0.09 [7.4] [4.8] [2.5] 0.07 [1.9] 0.84 0.84 0.49 -0.24 [19.8] [19.7] [15.4] [-3.6] 1.53 [20.4] 0.18 0.18 0.10 0.30 Column (6) & Column (10) 0.15 [1.89] 0.05 [2.53] 0.72 [19.62]
0.10 0.08 0.11 [8.6] [5.1] [8.8] 0.02 [2.4] 0.11 0.11 0.04 0.16 [8.6] [8.6] [3.6] [6.5] -0.04 [-2.3] 0.24 0.24 0.16 0.24
ce pt
FV
FV=CF_K Class 1 (6) (7)
ed
zit
Class 2 (3) (4)
us
FV=None Class 1 (1) (2)
Class 2: Not equity-dependent firms
Class 1 (13) (14) (15) (16)
Class 2 (17) (18) (19) (20)
FV=WorkingK_K Class 1 (21) (22) (23) (24)
Class 2 (25) (26) (27) (28)
0.20 0.19 0.21 0.17 0.16 0.16 0.21 0.22 0.19 0.18 0.17 0.16 [13.6] [13.5] [19.1] [13.9] [13.8] [19.1] [14.5] [14.3] [19.1] [16.8] [16.4] [19.1] 0.03 0.12 0.02 0.12 [1.3] [6.3] [0.8] [6.3] 0.01 0.01 0.23 0.02 0.01 0.01 0.04 0.00 0.09 0.09 0.51 0.10 0.02 0.01 0.05 0.00 [1.5] [1.4] [41.4] [2] [3.2] [2.8] [16.4] 0.00 [10.3] [10.2] [33.9] [10.5] [3.6] [3.2] [13.4] [0.6] 0.01 0.01 0.03 0.02 [2.4] [1.3] [2.7] [1.3] 0.22 0.22 0.41 0.22 0.29 0.30 0.24 0.29 0.25 0.25 0.32 0.25 0.29 0.30 0.21 0.29 Column (14) & Column (18) Column (22) & Column (26) 0.03 [2.03] 0.05 [1.76] -0.10 [-4.34] -0.11 [-4.5] 0.00 [4.97] 0.07 [9.79]
M an
Financial Variables
cr
Table 8 (cont'd) -Response of investment to alternative fundamentals and financial variables (when firms are grouped by their KZ index)
0.14 0.08 -0.58 [4.7] [2.1] [-18.9] 0.09 [2.8] 0.25 0.25 0.23 0.02 [42.8] [42.8] [41.4] [2.9] 0.29 [38.3] 0.46 0.46 0.41 0.66 Column (14) & Column (18) 0.01 [2.75] 0.07 [2.53] 0.22 [18.85]
0.09 0.07 0.12 [8.3] [4.9] [8.7] 0.02 [2.2] 0.04 0.04 0.04 0.05 [14] [13.9] [16.4] [11.2] -0.01 [-3.4] 0.27 0.27 0.24 0.27
0.21 [6.1]
0.13 -0.32 [3.2] [-10.7] 0.10 [2.8] 0.57 0.56 0.51 -0.08 [34] [34.1] [33.9] [-3.7] 0.87 [37.5] 0.36 0.36 0.32 0.60 Column (22) & Column (26) 0.06 [4.61] 0.08 [3.58] 0.53 [28.9]
0.09 0.07 0.10 [7.9] [4.6] [7.7] 0.02 [2.3] 0.04 0.04 0.05 0.05 [9.5] [9.5] [13.4] [5.4] -0.01 [-1.2] 0.24 0.24 0.21 0.24
Ac
Note: The estimation technique is OLS. Both the time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), zit is the mandated investment rate, and qit is Tobin's q. FV stands for a financial variable. There are 3 different financial variables used in the analysis: CF_K stands for the cash-flow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital. The definitions of the variables are given in Section 4.1. Class 1 (Class 2) firms are defined as the ones taking place in the higher (lower) 50 percentile of the firms when they are sorted by the size of their KZ index.
58 Page 58 of 60
ip t
ait
Size of capital Firms with small capital stock (1) stock Firms with large capital stock (2)
rait
zit
-2 stdev -1 stdev +1 stdev +2 stdev -0.545 -0.270 0.281 0.557
-2 stdev -1 stdev +1 stdev +2 stdev -0.570 -0.277 0.309 0.602
us
-2 stdev -1 stdev +1 stdev +2 stdev -0.410 -0.205 0.204 0.408
Shock values
cr
Table 9 - Joint effects of linear and nonlinear terms of fundamentals in investment equations
-0.042
0.056
0.126
-0.044
-0.028
0.043
0.097
-0.088
-0.050
0.070
0.151
(1)-(2)
-0.015 -0.056
-0.019 -0.023
0.041 0.015
0.106 0.021
-0.016 -0.028
-0.018 -0.010
0.038 0.004
0.096 0.001
-0.083 -0.006
-0.045 -0.005
0.060 0.011
0.126 0.026
Number of employees
Firms with low number of employees (3) Firms with high number of employees (4) (3)-(4)
-0.073 -0.008 -0.066
-0.043 -0.017 -0.026
0.055 0.044 0.011
0.122 0.114 0.008
-0.043 -0.011 -0.033
-0.028 -0.015 -0.013
0.043 0.037 0.006
0.098 0.094 0.004
-0.092 -0.073 -0.019
-0.050 -0.041 -0.009
0.069 0.059 0.010
0.146 0.127 0.020
Dividend payout ratio
Firms with low dividend payout ratio (5) Firms with high dividend payout ratio (6) (5)-(6)
-0.068 -0.021
-0.040 -0.024
0.052 0.050
0.116 0.126
-0.042 -0.027
-0.028 -0.022
0.043 0.041
0.099 0.099
-0.104 -0.048
-0.055 -0.034
0.069 0.063
0.143 0.146
-0.048
-0.017
0.002
-0.010
-0.015
-0.006
0.002
0.000
-0.056
-0.020
0.006
-0.003
Firms with low dividend to capital ratio (7) Firms with high dividend to capital ratio (8) (7)-(8)
-0.069 -0.015 -0.053
-0.039 -0.024 -0.015
0.048 0.056 -0.008
0.105 0.144 -0.039
-0.040 -0.024 -0.017
-0.027 -0.020 -0.008
0.044 0.037 0.007
0.102 0.089 0.013
-0.099 -0.062 -0.037
-0.053 -0.037 -0.016
0.070 0.056 0.014
0.147 0.124 0.023
Firms without bond rating (9) Firms with bond rating (10) (9)-(10)
-0.065 0.002 -0.067
-0.040 -0.009 -0.031
0.055 0.029 0.026
0.125 0.078 0.047
-0.042 0.024 -0.066
-0.028 0.001 -0.028
0.043 0.023 0.020
0.099 0.068 0.032
-0.091 -0.056 -0.035
-0.050 -0.033 -0.017
0.068 0.050 0.017
0.143 0.111 0.032
-0.057 -0.045
-0.038 -0.028
0.057 0.041
0.133 0.093
-0.041 -0.026
-0.030 -0.018
0.051 0.029
0.122 0.067
-0.106 -0.062
-0.056 -0.037
0.072 0.058
0.150 0.129
-0.012
-0.010
0.016
0.039
-0.014
-0.012
0.023
0.055
-0.044
-0.019
0.014
0.021
-0.080
-0.045
0.054
0.118
-0.098
-0.057
0.077
0.170
-0.101
-0.058
0.086
0.187
-0.016
-0.016
0.032
0.079
-0.007
-0.018
0.048
0.123
-0.005
-0.018
0.054
0.139
-0.064
-0.029
0.023
0.039
-0.091
-0.039
0.030
0.047
-0.096
-0.040
0.032
0.049
Mean
Standard deviation
Min
Max
Bond rating
Firms with high debt to capital ratio (11) Firms with low debt to capital ratio (12) (11)-(12)
KZ index
Equity-dependent firms (13) Not equity-dependent firms (14) (13)-(14)
Ac
ce pt
Total debt to capital ratio
ed
Dividend to capital ratio
M an
-0.071
Fundamentals ait rait zit
0.00 0.01 0.02
0.20 0.28 0.29
-0.69 -1.13 -1.13
0.70 0.94 1.14
25th 50th 75th percentile percentile percentile -0.12 -0.13 -0.14
0.00 0.01 0.02
0.11 0.15 0.16
Note: The gray-shaded rows are for the firms expected to be financially constrained. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), and zit is the mandated investment rate. stdev stands for standard deviation. Shock values are calculated as: (mean of fundamental +standard deviation of fundamental* degree of standard deviation) where the degree can be -2, -1, 1 or 2. The standard deviationas and means of each fundamental is given at the bottom of the table. The joint effects of linear and nonlinear terms of fundamentals are calculated as: b 1*shock value+b 2*(shock value)2 where b 1 and b 2 are the estimated coefficients from iit b1xit b2 xit2 cCF_ Kit T F uit , taken from columns (6) and (10) of Tables 2-8. In these calculations, the empirical specification includes cash flow to capital ratio as financial variable.
59 Page 59 of 60
Research Highlights
d
M
an
us
cr
ip t
Investment/fundamental sensitivity of firms with/without financial problems New fundamentals are used: profitability shocks and mandated investment rate Investment is regressed on fundamentals and financial variables for two groups Financially constrained firms are found to be more responsive to fundamentals Results support expectations of contracting models of financial market imperfections
Ac ce pt e
• • • • •
Page 60 of 60