Investment, Alternative Measures of Fundamentals, andFinancial Variables

Nihal Bayraktar∗,†

February 10, 2006

Abstract

The paper investigates the question of how successful financial variables are in ex-plaining firm-level fixed capital investment when investment opportunities are controlledfor by two of the recently-introduced empirical fundamentals: profitability shocks andthe gap measure between the desired and actual capital stocks (mandated investmentrate). Tobin’s q is also included in the analyses for the purpose of comparison. Thedata set, which is constructed from the COMPUSTAT database, includes U.S. basedmanufacturing firms. The results show that financial variables are important determi-nants of investment but they are not as significant as claimed by some empirical studiesfocusing on capital market imperfections. The explanatory power of financial variablesin the investment process declines with increasing significance of fundamentals. Anotherinteresting result is that the level of investment by expected-to-be financially constrainedfirms, identified by commonly used a priori measures of financial constraints, tends to berelatively less sensitive to changes in financial variables compared to changes in fundamen-tals even though the opposite is predicted in the literature. This result questions whetherinvestment-cash flow sensitivity can be a good measure of financial constraints, as well aswhether some of the firm characteristics used in identifying financially constrained firmsin the literature are sufficient.Keywords: investment, fundamentals, financial variables, financing constraints .JEL Classification Number : E22

∗Penn State University Harrisburg, School of Business Administration, Middletown, PA 17057, USA. E-mail: nxb23@psu.edu.

†I am grateful to John Haltiwanger and Plutarchos Sakellaris for their generous support and advice, andto John Shea for numerous comments and suggestions. Any errors are my own.

1

1 Introduction

Empirical fundamentals, such as average Tobin’s q, have produced disappointing empiricalresults in explaining the investment process of firms in the neoclassical investment literature.Even though the Q theory shows that a firm’s marginal q should be the only determinantof investment, a well-developed literature shows that investment is sensitive to a firm’s in-ternal funds after controlling for Tobin’s q, the ratio of asset market value to replacementcost of capital. Possible explanations for this high investment-internal fund sensitivity areinvestigated in two groups: the presence of financial market imperfections or the existenceof measurement problems related to average Tobin’s q, which prevent it from fully capturinginvestment opportunities.

In the first group, the financial market imperfections literature assumes that firms’ networth determines their financial position. When a firm’s net worth is low, this firm can beconsidered financially constrained since it is likely to face an asymmetric information problemin financial markets, which makes it difficult for them to find enough external funds to financetheir investment projects. Even if they could find external funds, they would be too expen-sive compared to the opportunity cost of internal funds.1 Because of this, their investmentdecisions are expected to be highly correlated with their internal funds. Firms with high networth, on the other hand, are expected to have a smaller asymmetric information problem;thus, they can find enough external funds to finance their capital adjustment, and follow theinvestment process suggested by changes in fundamentals. This implies that the investmentdecision of firms with high net worth would be independent of the availability of their inter-nal funds. Indeed, the empirical literature investigating financial market imperfections showsthat firms that are classified as financially constrained using a priori criteria present a largersensitivity of investment to internal funds even after investment opportunities are controlledfor by fundamentals, where the most commonly used one is Tobin’s q.2

In the second group of studies, the presence of measurement error problems related tofundamentals is given as an alternative explanation for why financial variables would ap-pear empirically more significant than fundamentals in determining investment. The lack ofimportance of Tobin’s q in determining investment can be caused by the possibility that itdoes not properly capture the quality of firms’ investment opportunities. They argue thatwhen measurement errors are controlled for, fundamentals become significant determinants

1 In the literature, it has been shown that the premium on external finance varies inversely with the firm’snet worth such that a fall in net worth causes the premium on external finance to increase, which may leadto a reduction in investment. Some examples of these studies are: Bernanke and Gertler (1990), Bernanke,Campbell and Whited (1990), Whited (1992), Hu and Schiantarelli (1998), Gilchrist and Himmelberg (1998),and Jaramillo, Schiantarelli, and Weiss (1996).

2See Kashyap, Lamont and Stein (1994), Carpenter, Fazzari and Petersen (1998), Hoshi, Kashyap, andScharfstein (1991) among others. See Schiantarelli (1996) and Hubbard (1998) for surveys.

2

of investment. Thus, as long as proper measures of investment opportunities are introduced,cash flow or other financial variables are not expected to add any new information in reducedform investment regressions.3 New measures of fundamentals are introduced in the literatureto overcome measurement error problems. For example, Gilchrist and Himmelberg (1995 and1998) introduce a “Fundamental Q” measure, which is the present discounted value of futureprofit rates. They show that investment is more sensitive to this fundamental compared toTobin’s q, but financial variables are still significant determinants of investment.

Recent papers based on investment models with non-convex adjustment costs have alsointroduced alternative empirical measures of fundamentals, two of which are profitabilityshocks and the gap measure between the desired and actual capital stocks (mandated in-vestment rate).4 Given these recently-introduced alternative empirical fundamentals, whichare expected to better capture investment opportunities than Tobin’s q, the question inves-tigated in this paper is whether the relative significance of firms’ financial position in theinvestment process may change when investment opportunities are controlled for by thesenew fundamentals.5 The answer to this question helps us better understand the relationshipbetween fundamentals and investment, and between financial variables and investment, bothof which are investigated extensively in the investment literature. The aim is to shed lighton the extent to which the investment-financial variable sensitivity can be linked to capitalmarket imperfections versus mismeasured fundamentals.

The analyses in this paper are based on a reduced form investment equation, in whichboth fundamental determinants of investment and financial variables are taken as explanatoryvariables. A panel data set at the firm level is constructed from the COMPUSTAT database.The data set includes U.S. manufacturing firms for the period of 1983-1996. The fundamentaldeterminants of investment are represented by profitability shocks and the gap measure.6

Tobin’s q is also included for the purpose of comparison. Financial variables are representedby the ratio of cash flow to capital, sales to capital, and working capital to capital.

The empirical results show that financial variables are still important determinants of

3Kaplan and Zingales (1995 and 1997), Gomes (2001), Erickson and Whited (2000), Cooper and Ejarque(2003), and Abel and Eberly (2003) study this issue.

4See Caballero, Engel and Haltiwanger (1995), Cooper and Haltiwanger (2005), and Cooper and Ejarque(2003). These fundamentals are compared with "Fundamental Q" and Tobin’s q in Bayraktar (2002). They arefound to be more significant in explaining investment compared to Tobin’s q and “Fundamental Q.” Bayraktar,Sakellaris, and Vermeulen (2005) show that financial variables are also important in determining investmentin addition to fundamental determinants, using a structural investment model based on both convex andnon-convex adjustment costs, where fundamentals are measured by profitability shocks.

5Despite the fact that the new fundamentals are constructed using current variables, they present a forward-looking behavior.

6A "Fundamental Q" measure calculated by Gilchrist and Himmelberg (1995, 1998) is not included sincethey have already reported that the signifinance of financial variables drops when investment opportunitiesare captured by "Fundamental Q."

3

investment, but they are not as significant as claimed by studies focusing on capital marketimperfections. The findings indicate that the explanatory power of financial variables inthe investment process declines with increasing significance of fundamentals. On the onehand, the explanatory power of financial variables in a reduced form investment equation isthe lowest when investment opportunities are measured by the gap between the desired andactual capital stocks.7 Tobin’s q, on the other hand, is the weakest determinant of investment,and financial variables have the highest explanatory power for investment when Tobin’s q isthe proxy for investment opportunities. Thus, this result implies that the previous empiricalfailure of fundamentals against financial variables might be caused by measurement errorsin fundamentals. When investment opportunities are captured better, the statistical andeconomic significance of financial variables drops.

Similar analyses are repeated after firms are classified into different groups using six al-ternative a priori criteria, which are commonly used in the capital market imperfectionsliterature to identify financially constrained firms. The firm characteristics used in deter-mining financially constrained firms are the level of capital stock, number of employees,dividend to capital ratio, dividend payout, debt to capital ratio, and firms’ bond rating.The empirical results based on these sub-samples report how the response of investment tofundamentals, and to financial variables changes, depending on whether firms belong to afinancially constrained or relaxed group. It is expected that firms with financial constraintsexhibit a significant investment-cash flow sensitivity than firms that appear less financiallyconstrained. An interesting result is that when profitability shocks and the mandated invest-ment rate are the fundamentals, the sensitivity of investment to financial variables, such ascash flow to capital ratio, tends to be lower for expected-to-be financially constrained firmseven though the opposite is expected in the literature. However, when Tobin’s q is the funda-mental measure, the results are as expected in the literature, such that financial variables aremore important in determining investment for firms taking place in a financially constrainedgroup. One implication of this result is that a high investment-cash flow sensitivity maynot be seen as evidence of financial constraints. A high sensitivity of investment to finan-cial variables for financially constrained firms might be caused by the fact that investmentopportunities are captured by insufficient measures of fundamental such as Tobin’s q.8 Theresults also imply that a priori criteria used in classifying firms may not be that successful inidentifying financially constrained ones.

The rest of the paper is organized as follows. Section 2 gives information about the

7As investigated in Bayraktar (2002), this fundamental measure is the most significant empirical determi-nant of investment when compared with other fundamentals.

8Kaplan and Zingales (1995 and 1997) also indicate that firms classified as less financially constrainedexhibit a significantly greater investment-cash flow sensitivity than firms classified as more financially con-strained.

4

relationship between investment, fundamentals, and financial variables. In Section 3, detailson the data set and variables are given. In Section 4, the empirical results are presented.Section 5 concludes.

2 Investment, Fundamentals, and FinancialVariables

The Q theory of investment presents a formal link between a firm’s investment and marginalq, and shows that marginal q should be the sole determinant of investment. This result canbe illustrated using the following neoclassical model with convex capital adjustment costs.The purpose of the competitive firm’s manager is to maximize the present discounted valueof the firm:

V (Ait,Kit) = maxIit

Π(Ait,Kit)−C(Kit, Iit) + βEAit+1|AitV (Ait+1,Kit+1), (1)

subject to the following constraint:

Iit = Kit+1 − (1− δ)Kit,

where the subscripts i and t denote the firm level variables and time period, respectively. V (·)is the value function, βEAit+1|AitV (·) is the present discounted future value of the firm, C(·)is the investment cost function, Iit stands for investment, Kit is the current capital stock,and δ is the depreciation rate. Ait is the profitability shock in period t. Π(·) is the profitfunction. β is the fixed discount factor.

It is assumed that both C(·) and Π(·) are homogenous of degree one in investment andcapital (homogeneity assumptions). C(·) is assumed to be a convex function such that:

C(Kit, Iit) = pIit +γ

2

∙IitKit

¸2Kit. (2)

Given these assumptions, we can scale equation (1) by Kit:

v(Ait) = maxiit

π(Ait)− c(iit) + β(1− δ − iit)EAit+1|Aitv(Ait+1), (3)

where iit = Iit/Kit, v(Ait) = V (Ait,Kit)/Kit, π(Ait) = Π(Ait,Kit)/Kit, c(iit) = C(Kit, Iit)/Kit,and β(1− δ − iit)EAit+1|Aitv(Ait+1) = βEAit+1|Ait

V (Ait+1,Kit+1)/Kit.

Maximizing equation (3) gives the following first order condition:

∂c(·)∂iit

= βEAit+1|Aitv(Ait+1). (4)

5

From equation (2),∂c(·)∂iit

= p+ γ.iit. (5)

If we combine equations (4) and (5), the investment rate is going to be a function ofmarginal q

iit =1

γ(βEAit+1|Ait

v(Ait+1)− p), (6)

where βEAit+1|Aitv(Ait+1) is marginal q. Given homogeneity assumptions listed above, it can

be shown that marginal q is equal to average q.

βEAit+1|Aitv(Ait+1) = βEAit+1|Ait

V (Ait+1,Kit+1)/Kit+1,

where average q is V (Ait+1,Kit+1)/Kit+1. Since marginal q is not empirically observable, itis replaced by average q (Tobin’s q) in empirical studies. The explanatory power of Tobin’sq for investment has not only been found to be negligible, but they have also producedextremely high parameter estimates for capital adjustment cost functions, which is γ in themodel presented above.

The empirical failure of neoclassical investment models has led to a search for alternativedeterminants of investment in the literature. One of the most prominent group of studiesfocuses on the importance of the financial position of firms in explaining their investmentbehavior. It is theoretically assumed that firms’ net worth determines their financial posi-tion, which, in turn, determines their investment behavior. Firms with low net worth areconsidered financially constrained since they are likely to face an asymmetric informationproblem in financial markets, which prevents them finding cheap external funds to financetheir investment projects. In this case, investment are expected to be highly correlated withinternal funds. Firms with high net worth, on the other hand, are expected to have a smallerasymmetric information problem; thus, they can borrow external funds, and follow the in-vestment process suggested by fundamentals, independent of the availability of their internalfunds.

It has been empirically shown that investment is sensitive to internal funds after control-ling for q, using a reduced form investment equation, which can be written, in general terms,as follows:

iit = bxit + cFVit + Tα+ Fφ+ uit, (7)

where i represents firms and t represents years. iit is the investment rate at firm i in period t.While x is a fundamental measure used in capturing investment opportunities, FV capturesfinancial variables such as net worth or internal funds. T represents a set of time dummies,and F represents a set of firm dummies used in removing fixed effects. u is the error term.This regression equation estimates the sensitivity of investment to changes in internal funds,

6

FV , controlling for investment opportunities, x. If fundamentals are the sole determinantsof investment, as specified by the Q theory, the coefficient of FV should be statisticallyinsignificant, given that fundamentals are not mismeasured. In the literature, the equation isestimated separately for financially constrained and unconstrained firms, which are identifiedby a priori proxies. The purpose is to distinguish between possibly mismeasured fundamentalsand capital market imperfections.9 It has been shown that the sensitivity of investment tofinancial variables is higher for financially constrained firms, which has been taken as evidencein favor of capital market imperfections.

There are three basic issues in the financial market imperfections literature: determininggood proxies for internal funds, identifying financially constrained firms, and capturing in-vestment opportunities by variables expected to be free of mismeasurement problems. Thecash-flow-to-capital ratio is the most commonly used financial variable to capture changes ininternal funds in the investment equation. One would expect that financial constraints shouldmake investment more responsive to this ratio. Even though it is used extensively in the ex-isting literature, its ability to capture the financial position of firms is questionable. Thebasic problem is that cash flow may contain information on future profits as well as on firms’financial position. In addition, cash flow may not be a good measure of changes in firms’net worth. Because of this, alternative financial variables, which are hopefully less correlatedwith investment opportunities, have been introduced. One alternative is stock measures ofinternal funds since they are less directly linked to investment opportunities. A commonlyused candidate is the cash-and-equivalents-to-capital ratio, which captures the short-termliquid asset position of firms. Another stock measure is the ratio of working capital (currentassets minus current liabilities plus inventories) to capital, which captures the leverage po-sition of firms net of current liquid assets (Gilchrist and Himmelberg, 1998). Tax paymentscan also be an instrumental variable for cash flow (Hubbard, Kashyap, and Whited, 1995).The sales accelerator investment demand literature claims that the availability of internalfunds depends on sales; investment therefore should be responsive to fluctuations in the salesratio. Thus, the sales-to-capital ratio is another variable used by the existing literature.

With regard to identifying financially constrained firms, the level of dividend to incomeratio is a commonly used indicator. Fazzari, Hubbard, and Petersen (1988) group firms ac-cording to this a priori measure of financial constraints, and compare the investment-internalfunds sensitivity of different groups. More specifically, they indicate that if information prob-lems in capital markets cause financing constraints on investment, this should be most clearlyseen for firms that retain most of their income. Thus, one would expect that the lower thedividend payout ratio is, the more financially constrained firms are. The reason for this ex-pectation is that if firms pay a high dividend relative to their income, they reveal that the

9Fazzari, Hubbard, and Petersen (1988) is the first paper following this methodology.

7

opportunity cost of their internal funds is low, so they are financially unconstrained. Thesize of firms in terms of their capital stock, total assets, or the number of employees is alsoused in selecting financially constrained firms, where smaller firms are expected to be morefinancially constrained. One explanation is that external funds have a significant fixed costcomponent creating increasing returns, thus, smaller firms have to pay higher costs to obtainexternal costs. Another restriction on the availability of external funds for smaller firms isthat public information about their investment projects is generally more limited. Another setof variables used in identifying financially constrained firms is their debt structure. Whited(1992) and Calomiris, Himmelberg, and Wachtel (1994) group firms depending on whetherthey have bond ratings or not. They argue that public debt issuance is a good indicator offirms’ financial position since it provides a low-cost access to capital markets. Firms’ stockor flow debt burden might be another indicator of financial problems, where debt stock isdefined as the market value of debt over the total market value of firms, and debt flow is de-fined as interest expense over current asset (Bernanke, Campbell, and Whited, 1990). Otherpossible measures of debt burden are interest coverage, which is interest expenditure overinterest expenditure plus cash flow (Whited, 1992), and the ratio of liquid assets to capital(Hu and Schiantarelli, 1998). One would expect that the larger the debt burden is, the morefinancially constrained firms are since they are expected to pay higher premiums to obtainexternal funds. Not only are many different indicators used in identifying firms with financialproblems, but their cut off values also differ across studies, which use the same indicator todetermine financially constrained firms.10

Despite the presence of a large range of financial variables used in determining firms’financial position, most empirical studies use only average Tobin’s q as a proxy for investmentopportunities.11 Theoretically, the correct measure of capturing investment opportunities ismarginal q, which is defined as the present discounted value of future profits generated by anadditional unit of capital. Since marginal q is not empirically observable, different substitutesare introduced in empirical studies such as Tobin’s q, defined as the ratio of firm’s averagevalue to its capital stock. Empirical results show that the explanatory power of Tobin’s qin investment equations is much weaker compared to financial variables, where a possible

10For example both Fazzari, Hubbard, and Petersen (1988) and Gilchrist and Himmelberg (1998) use thedividend payout ratio in splitting their samples. The first study, on the one hand, splits the sample of firmsinto three different groups, depending on whether the ratio is less than 0.1, between 0.1 and 0.2, or largerthan 0.2. The second study, on the other hand, splits the sample in two, and defines the group of financiallyunconstrained firms as those in the top one third of the dividend payout ratio.11One of the exceptions is Gilchrist and Himmelberg (1995 and 1998). They introduce "Fundamental

Q" as s new fundamental, and investigate the effectiveness of this new fundamental measure in explaininginvestment, and the role of financial variables in this process. They show that even though financial variablesare still statistically significant determinants of investment, the new fundamental measure is more successfulin explaining investment than Tobin’s q.

8

reason for this failure would be inadequacy of average Tobin’s q in capturing investmentopportunities due to measurement errors.12 Gomes (2001) argues that if fundamentals aremeasured accurately, there is no reason for financial variables to be significant determinantsof investment even if financial constraints are present, since information on the financialposition of firms is expected to be already included in fundamentals. A recent investmentliterature based on non-convex capital adjustment cost models introduces new measures offundamentals, two of which are profitability shocks and the gap measure between the desiredand actual capital stock.13 In this paper, these two fundamentals are taken as proxies forinvestment opportunities to better understand whether sensitivity of investment to financialvariables can be explained by measurement errors in fundamentals. One advantage of thesefundamentals is that, even though they are constructed using current variables, they presenta forward-looking behavior. Since profitability shocks are serially correlated, the currentvalue of shocks gives information about future profitability. The gap measure between thedesired and actual capital stocks is also informative about future investment behavior sincethe magnitude of the gap determines whether a firm invests in the current period or in thefuture. Because of this, the correlation between these fundamentals and internal funds islow.14

3 Data and Variables

The main data source is the COMPUSTAT firm-level database. The data set, covering theperiod from 1983 to 1996, includes U.S. manufacturing firms.15 The total number of firmsis 463 and the total number of panel observations is 6450.16 The following sub-sectionsintroduce main variables used in the study.

12Erickson and Whited (2000), using measurement error—consistent generalized method of moments estima-tors, also find that investment-cash flow sensitivity might be reasoned by measurement error. Also see Kaplanand Zingales (1995 and 1997), Cooper and Ejarque (2003), and Abel and Eberly (2003).13See Caballero, Engel, and Haltiwanger (1995) and Cooper and Haltiwanger (2005). As investigated in

Bayraktar (2002), these measures are more successful in explaining investment than Tobin’s q.14For example, the correlation between the profitability shocks and the cash flow to capital ratio is only

0.25 as shown in Bayraktar (2002).15Since the retirement data, used in constructing investment series, were not collected since 1996, the

following years are not included in this study.16The balanced data set would have had 6482 observations. Detailed information on the data set and

variables is given in Appendix A.

9

3.1 Investment

The definition of capital includes plant, property, and equipment, and investment is definedas capital expenditure net of capital sales, including capital retirements.17 The replacementvalue of capital is calculated using a perpetual inventory method as follows:

Kt = (1− δ)Kt−1 + It, (8)

where Kt is the real capital stock, It is real investment, which is calculated by deflatingthe nominal value by the 4-digit investment price index. δ is the 2-digit depreciation ratefrom the Bureau of Labor Statistics (BLS) database. It is equal to the average value ofthe depreciation rates for the period of 1981-1996. The investment rate is defined as theratio of real investment to the replacement value of capital. The distribution function ofthe investment rate is presented in Figure 1. Since investment is net of sale of capital,there are negative investment rates available, corresponding to nearly 10 percent of the totalobservations. Descriptive statistics are given in Table 1.

3.2 Financial variables

The following financial variables are used in capturing the effects of internal funds on invest-ment:18

• Cash flow to capital ratio: The ratio of the book value of cash flow to the beginning ofperiod book value of gross total plant, property, and equipment (PPE).

• Sales to capital ratio: The ratio of the book value of net sales revenue to the beginningof period book value of PPE.

• Working capital to capital ratio: The ratio of the book value of working capital (thedifference between current assets and liabilities) to the beginning of period book valueof PPE.

The following a priori proxies are used in identifying financially constrained firms:

• Book value of PPE stock: Small firms are expected to be financially constrained.17As defined in the COMPUSTAT User Guide, the sale of capital and retirements data are combined for

some firms, but they are separate series for others. In order to obtain a uniform series, the retirements dataand sale of capital data are added up whenever the sale of capital data have a lower value than the retirementsdata, indicating that retirements are not included in the sale of capital data.18Detailed explanations for these variables are presented in Appendix A.2.

10

• Number of employees: Firms with a less number of employees are expected to be finan-cially constrained.

• Total debt to capital ratio: This is defined as the ratio of the book value of total debt tothe beginning of period PPE stock. Firms with a high debt to capital ratio are expectedto be financially constrained.

• Bond rating: This is the issuer senior debt rating assigned by Standard & Poor’s. Firmswith a debt rating are expected to be less financially unconstrained since they can findexternal funds more easily.

• Dividend payout : This is defined as the ratio of the book value of dividends to the bookvalue of income before extraordinary items. When this ratio is low, firms are morelikely to be financially constrained.

• Dividend to capital ratio: This is the ratio of the book value of dividends to the begin-ning of period book value of PPE. The lower the ratio is, the more constrained firmsare expected to be.

3.3 Fundamentals

The purpose of this study is to assess the explanatory power of internal funds when firms’investment opportunities are proxied by the mandated investment rate and the profitabilityshocks, where the last one is calculated following two different ways. Besides these funda-mentals, Tobin’s q is also included for the purpose of comparison.

3.3.1 The Mandated Investment Rate

Caballero and Engel (1994) try to explain the lumpy nature of investment using a model basedon the standard (S,s) literature. They measure imbalances in capital as the gap betweenthe desired and actual capital stocks (the mandated investment rate). In their model, theinvestment rule is that once the measure of imbalance reaches a threshold value, the capitaladjustment occurs at once. The reason for firms to wait until they reach the trigger point isexplained by the presence of non-convex adjustment costs, describing an increasing returnson capital adjustment technology. This model is empirically studied by Caballero, Engel, andHaltiwanger (1995).19 They show that the response of investment to the gap is nonlinear,supporting the availability of non-convex capital adjustment costs.

19The mandated investment rate is also used in explaining the fixed capital investment process by Goolsbeeand Gross (1997).

11

The mandated investment rate, xit , is defined as the deviation of the desired capital stockfrom the actual one such that

xit ≡ ekit − kit−1, (9)

where ekit and kit−1 represent the natural log of desired and actual capital stocks in firm i attime t. While positive values of xit indicate capital shortages, negative ones indicate excesscapital. Desired capital refers to the stock of capital that a firm would hold if adjustmentcosts were momentarily removed. It is constructed under the following assumptions. First,it is assumed that desired capital is proportional to the log of frictionless capital stock, k∗it,such that ekit = k∗it + di, (10)

where di is a firm-specific constant.20 Frictionless capital is the stock of capital that a firmwould hold if it never faced adjustment costs.

The second assumption is that frictionless stock of capital, k∗it, is determined by a neo-classical model. This model produces the following empirical equation for the gap measure:21

k∗it − kit−1 = ηi{(yit − kit−1)− ψicit}, (11)

where yit and cit represent the natural log of the value of output and cost of capital in firmi at time t, respectively. ηi is assumed to be equal to (1/(1− α)) where α is the cost shareof capital. ψi is the long-run elasticity of capital with respect to its cost.

The third assumption is the estimation of ψi from a cointegrating regression of the nat-ural log of the capital-to-output ratio on the cost of capital at the 2-digit industry level usingfirm-level panel data.22 This coefficient can be interpreted as the long-run elasticity of cap-ital with respect to its cost. Bayraktar (2002) calculates the average value of this measureas -0.68. This value is close to -1, which is the long-run elasticity in neoclassical models.After calculating frictionless capital, the firm-specific constant, di, is estimated by takingthe average gap between kit−1 and k∗it for the five points with investment closest to medianinvestment, which can be thought as maintenance investment.

3.3.2 Profitability Shocks

The second variable capturing investment opportunities in this paper is the idiosyncraticprofitability shock. Cooper and Haltiwanger (2005) display that the empirical relationship

20Bertola and Caballero (1994) present that this assumption is compatible with the behavior of a rationalfirm whose profit function is isoelastic, and that faces shocks that have independent increments.21Details are given in Appendix B2.22The twice-lagged first difference of the log of cost of capital is included as explanatory variable in order

to reduce the small sample bias.

12

between the investment rate and profitability shocks is nonlinear and asymmetric such thatthe response of investment to positive shocks is much stronger than its response to negativeshocks. They show that this behavior of investment is better explained when both convexand nonconvex adjustment costs are taken into account together in a model. The profitabilityshocks can be presented in the following firm-level profit function:

Π(Ait,Kit) = AitKθit, (12)

where Ait is the profitability shock, which consists of both aggregate and idiosyncratic com-ponents, θ is the curvature of the profit function, and Kit is the firm level capital stock. θ isestimated by regressing the natural log of net profit (net of cost of production) on the log ofthe replacement value of capital stock using firm-level panel data.23

There are two alternative ways of calculatingAit. The first way calculatesAit by regressingthe log of profits on the log of real capital. Time dummies are included to remove the effectsof aggregate profitability shocks.

lnΠ(¦) = θ. lnKit + Fα+ Tφ+ rait,

where rait is the error term taken as idiosyncratic profitability shocks, named as residualprofitability shocks from now on.

The second way calculates Ait indirectly through the first order condition for profit max-imization with respect to employment. Since the employment series are more reliable, thesecond way allows us to avoid possible measurement errors in profit data. In this case, Ait isgoing to be a function of the shocks to revenue, cAit, and other parameters such that:24

Ait = f(cAit, ξ, αK , αL, w),

where ξ the price elasticity of demand, αK and αL are the shares of capital and labor costs,respectively, and w is the wage level. The aggregate shocks are calculated as the annual meanof Ait, and the idiosyncratic component of Ait, presented by ait, is taken as the deviation fromthis mean. ait is named, from now on, as profitability shocks from the first order condition.25

3.3.3 Tobin’s q

Tobin’s q (average 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.26 There23θ is assumed to be the same for each firm in each period. However, if there are structural differences

across firms, they need to be removed from affecting the analysis. Consequently, we remove fixed effects tosolve a possible structural heterogeneity problem.24Details on calculation of these shocks are given in Appendix B.3.25The correlation coefficient between rait and ait is given as 0.55 in Bayraktar (2002).26Hayashi (1982) and Abel (1979) report that the neoclassical model with convex adjustment costs yields

a q value, which is known as marginal q. Since it is not empirically feasible to calculate this marginal value,

13

are many different ways available to measure the market value of firms and the replacementvalue of their capital stocks. In this study, the definition of Tobin’s q is the one used byBarnett and Sakellaris (1998).27 The numerator is the sum of the market value of commonstock, the liquidating value of preferred stock, the market value of long-term debt and thebook value of short-term debt. The denominator, on the other hand, is the sum of thereplacement value of fixed capital and inventories.

4 Empirical Results

In this section, the following reduced form investment equation is estimated using a leastsquares regression technique for panel data:

iit = b1xit + b2x2it + cFVit + Tα+ Fφ+ uit,

where i represents firms and t represents years. iit is the investment rate at firm i in periodt. While xit is a fundamental measure capturing investment opportunities, FV capturesfinancial variables such as net worth or internal funds. T represents a set of time dummies,and F a set of firm dummies to remove fixed effects. u is the error term. In this equation, fourdifferent fundamentals are included: two types of profitability shocks, the gap between thedesired and actual capital stock (mandated investment rate), and Tobin’s q. Since Bayraktar(2002) shows that the relationship between the investment rate and these fundamentals isnonlinear, the squared term of fundamentals is also included.28 In this study, the cash flowto capital ratio, the ratio of sales to capital, and working capital to capital are included asfinancial variables.29 In addition to full sample analyses, the sample is split into sub groupsusing six alternative a priori criteria to identify financially constrained firms. The criteriaintroduced in this study are the size of capital stock, number of employees, dividend to capitalratio, dividend payout ratio, debt to capital ratio, and firms’ bond rating.30

the average value of q can be used as a proxy for the marginal value under some strict assumptions, as shownin Section 2. The assumptions are that firms have a linear homogeneous net revenue function, and operate inperfectly competitive markets.27Details on the calculation of these variables are given in Appendix B.1.28Related to the nonlinear relationship between fundamentals and investment, also see Barnett and Sakellaris

(1998), Barnett and Sakellaris (1999), and Abel and Eberly (2002).29All 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 tocapital and the ratio of tax payments to income are also included. But the results are not reported becausethe ones with the cash and equivalence to capital ratio were similar to the results with the ratio of cash flowto capital, and the statistical significance of tax payments was negligible.30The definitions of these variables are given in Section 3.2. Besides them, other criteria such as total assets,

real capital stock, flow and stock of debt burdens, and interest coverage ratio are also introduced. But, theresults are not reported in the paper since they produce similar results.

14

In the following sections, the full sample results are reported first, then the sub-sampleresults are presented. The last set of results are presented in three groups: for firms groupedaccording to the size of their capital stocks and the number of employees; for firms groupedaccording to their dividend ratios; and for firms grouped according to their debt stock andthe availability of bond rating.

4.1 Full sample results

The estimated coefficients are reported in Table 2. Four sets of results, corresponding toeach fundamental determinant of investment, are presented in the table. In the first column,financial variables are excluded from the explanatory variable set. In the following columns,different financial variables are introduced. By comparing the results in the first columnand in the following columns, the marginal explanatory power of financial variables can beunderstood.

The estimated coefficients of both the linear and squared terms of the fundamentals,including Tobin’s q, are statistically significant at 1 percent. When we check the resultswithout financial variables, the highest adjustedR2, on the one hand, belongs to the regressionresult with the mandated investment rate. The regression result with Tobin’s q, on the otherhand, produces the lowest adjusted R2. Some of these results are comparable to the onesreported in previous studies. For example, Barnett and Sakellaris (1998) present regressionresults for Tobin’s q and its squared term. Even though their data set is also constructedfrom the COMPUSTAT database, their sample period and the definition of investment isdifferent. These differences in the data sets are reflected in results. While the estimatedcoefficient of Tobin’s q is 0.057 in Barnett and Sakellaris (1998), it is 0.19 in this paper. Onthe other hand, while the value of the coefficient of the squared term is -0.006 in their study,it is 0.043 in this paper. The results with the profitability shocks obtained from the firstorder condition (ait) can be compared to the results presented in Cooper and Haltiwanger(2002). Even though they use plant-level data in their analysis, the values of the estimatedcoefficients are close to each other. While their estimates are 0.316 and -0.041 for the leveland squared terms of shocks, respectively, these values are 0.208 and 0.010 in Table 2.

The results with the cash flow ratio (CF_K) are reported in the second column of thetable. This variable is statistically significant in each regression except when the mandatedinvestment rate (kit) is the fundamental measure. This means that the sensitivity of in-vestment to CF_K is negligible when investment opportunities are controlled for by themandated investment rate. The other interesting result is that while the inclusion of CF_Kdoes not much change the value of the adjusted R2 in the first three sets, it jumps from0.056 to 0.134 when Tobin’s q is the fundamental measure. The sensitivity of investment toCF_K is also considerably higher in the last case. Since the cash flow-to-capital ratio is one

15

of the most commonly used financial variables in investment regressions, there are severalcomparable studies in the literature. For example, Fazzari, Hubbard, and Petersen (1988)find that the estimated coefficient of the cash flow ratio is approximately around 0.52, wherethey include only the level of Tobin’s q to capture investment opportunities. The estimatedvalue of the cash flow ratio in Table 2 is 0.48 when Tobin’s q is the fundamental measure andits squared term is included as well.

When the effect of financial variables are captured by the ratio of sales to capital (Sales_K),we obtain similar results, as reported in column 3. The basic difference is that the estimatedcoefficient of Sales_K is also statistically significant when the mandated investment rate isthe fundamental variable. Again the sensitivity of investment to this financial variable getsthe highest value when Tobin’s q is the fundamental measure. The ratio of working capitalto capital (WorkingK_K) contributes to the explanation of investment in a statisticallysignificant way in each equation. As is the case in other results, the sensitivity of investmenttoWorkingK_K is the highest with Tobin’s q and the lowest with the mandated investmentrate.

Overall, even though financial variables are still significant determinants of investment,the results show that the explanatory power of financial variables drops when investmentopportunities are controlled for by the new fundamentals. As discussed in Section 2, theresults support the idea that the significance of financial variables in determining investmentmay be reasoned by mismeasured fundamentals, which prevent them capturing investmentopportunities successfully.31

4.2 Results Based on Sub-samples

In the following sub-sections, the question of how the relationship between investment, funda-mentals, and financial variables changes when firms are grouped into sub-samples accordingto their financial positions, using a priori criteria extensively observed in the literature. Thesedifferent criteria provide alternative definitions of financially constrained status. The resultsare collected in three groups.

4.2.1 Size of Firms

The first two criteria are the size of firms in terms of their capital stocks and the number ofemployees. Small firms are defined as the ones with the average capital stock (or the numberof employees) in the bottom half of the empirical distribution. While Class 1 stands for

31For example, using measurement error—consistent generalized method of moments estimators, Erikson andWhited (2000) show that most of the stylized facts produced by investment, q, and cash flow regressions arereasoned by measurement error. Cash flow also does not matter after controlling for measurement errors.

16

the group of small firms, Class 2 is the group of large firms. Since firms with a low capitalstock typically have few employees, the results obtained using these two different criteria aresimilar. One would expect smaller firms to be financially constrained, since costs of gettingexternal funds are presumably higher for them, and public information about their investmentprojects is generally more limited. These facts restrict their ability to find external funds.Since small firms are expected to be financially constrained, their investment should respondless to changes in fundamentals, but more to internal funds.

The average values of variables for these two groups of firms are reported in the first twocolumns of Table 3. The average investment rate in both groups is similar. The growth rateof sales for small firms is 21 percent versus only 7 percent for large firms. This means thatsmall firms grow much faster. The earnings retention rate is higher for small firms, while thedividend payout ratio is lower. While the average value of Tobin’s q is much higher for smallfirms, indicating that these firms are supposed to invest more, the working capital ratio ofsmall firms is also higher. When the average values of fundamentals are compared in twogroups, they are closer for the profitability shocks and the mandated investment rate.

The estimation results for firms with small versus large capital stocks are presented inTable 4. In almost each case, while the coefficient of the linear term of the fundamentalsis larger for small firms, the coefficient of the squared term is lower. One possible reasonfor this result might be lower non-convex capital adjustment costs, such as fixed costs, forsmaller firms. As a result, these firms can linearly follow investment opportunities. In eachset, the adjusted R2 is lower for small firms, indicating that the explanatory power of thefundamentals and financial variables for investment is lower. For this group, this gap isrelatively larger in the results with Tobin’s q, where the adjusted R2 is 0.047 for small firms,and 0.263 for large firms, given that financial variables are excluded.

Another interesting observation is that when we compare the relative magnitudes of theestimated coefficients, large firms’ investment is more sensitive to changes in financial vari-ables even though these variables are expected to be more important for smaller firms inthe literature.32 The exception is Tobin’s q, for which the coefficients of financial variablesare higher for smaller firms as presented in many empirical studies. In the capital marketimperfections literature, smaller firms are identified as relatively more financially constrainedsince their investment is more sensitive to changes in internal funds. But this result is not

32The high sensitivity of investment to the cash flow ratio for financially unconstrained firms is also observedin the study by Kaplan and Zingales (1997). The least constrained and the most financially successful firms intheir sample seem to depend primarily on their cash flow to finance their investment despite the availability ofadditional low cost funds. In their study, they use Tobin’s q to control investment opportunities, and classifyfirms 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 arecorrected.

17

observed when investment opportunities are controlled for by new fundamentals. In fact,the coefficient on CF_K is negative but statistically insignificant for smaller firms when themandated investment rate is the fundamental measure. There might be different reasonsfor why less financially constrained firms exhibit higher investment-internal funds sensitiv-ity. As specified by Kaplan and Zingales (1997 and 2000), one possible reason could beexcessive conservatism by managers, which may be caused by the way firms are organized ornon-optimizating behavior of managers.

The average values of the variables, when firms are grouped according to the numberof employees are reported in the third and fourth columns of Table 3. Again smaller firmsgrow faster, retain a higher fraction of their income, and pay fewer dividends as a proportionof their income. The average investment rate is a bit higher for small firms, where it is 14percent for smaller firms on average, and 12 percent for large firms.

The estimation results for these two groups are presented in Table 5. As was the casein Table 4, the sensitivity of investment to the linear term of the fundamentals is higherfor smaller firms; but the sensitivity of investment to the squared term is higher for largerfirms. The adjusted R2 value is higher for larger firms in each case. The magnitude ofthe estimated coefficients on financial variables is higher for larger firms except when thefundamental measure is Tobin’s q. This indicates that results are robust whether the sampleis split by the size of capital stock or by the number of employees.

4.2.2 Size of Dividends

The third and the fourth criteria to identify financially constrained firms are the dividendpayout ratio and the dividend to capital ratio, respectively. Firms are classified as financiallyconstrained if the ratios are in the bottom half of the empirical distribution. One wouldexpect firms with a high dividend ratio to be financially unconstrained since, if they pay alarge fraction of their income as dividends, they reveal that the opportunity cost of theirinternal funds is low, thus financially relaxed. Since firms with a low dividend ratio areexpected to be financially constrained, their investment should respond less to changes infundamentals, but more to changes in internal funds. Class 1 is the group of firms with a lowdividend ratio, and Class 2 with a high dividend ratio. The average values of the variablesfor two groups are reported in the fifth and sixth columns of Table 3. The capital stock andthe number of employees are lower for Class 1 firms; they grow faster and retain a largerfraction of their income. While the average investment rate is 0.15 for Class 1 firms, it is 0.11in Class 2. The average value of Tobin’s q is higher for Class 1 firms.

The estimation results for these two groups of firms are presented in Table 6. Whilethe sensitivity of investment to the linear term of the fundamentals is higher for firms withlow dividend payout, the sensitivity to the squared term is lower. This means that, as was

18

the case for smaller firms, those with lower dividend payout follow fundamentals linearly,indicating a lower significance of non-convex capital adjustment costs. Even though Class 1firms are expected to be financially constrained since they have lower dividend payout, thesensitivity of the investment rate to cash flow ratio is lower compared to its sensitivity to thenew fundamentals. But when Tobin’s q is the fundamental measure, the estimated coefficientsof financial variables are always higher for firms with a lower dividend ratio, independent ofwhich financial variable is included in the regression equation. Thus, the results with Tobin’sq are more consistent with expectations in the literature.33 On the other hand, these firms’investment is more sensitive to the working capital ratio and the sales to capital ratio inalmost each case, independent of which fundamental variable is used. One possible reasonfor these results might be that firms paying low dividends are also have less cash flow; they,in turn, would depend on their working capital to finance their investment.

The average values of the variables for firms grouped according to their dividend to capitalratio are reported in Table 3. Since firms paying low dividends as a fraction of their incomealso usually have a low dividend to capital ratio, the results are similar. The estimationresults for these two groups of firms are presented in Table 7. The results are similar to thosein Table 6. The explanatory power of the financial variables for investment is lower for firmspaying low dividend as a fraction of their capital. When the new fundamentals are used inthe regression equations, the sensitivity of investment to financial variables is much higherwhen Tobin’s q is the fundamental measure.

4.2.3 Size of debt to capital ratio and bond rating

The fifth and the sixth criteria used in identifying financially constrained firms are the avail-ability of a bond rating and the size of the debt to capital ratio, respectively. Since the debtburden of firms with a high debt ratio is heavier, they are expected to be more financiallyconstrained, so their investment should be more sensitive to changes in internal funds. Sim-ilarly, firms without any bond rating are likely to be more financially constrained than theones with a bond rating since public debt issuance would be a good indicator of low-costaccess to capital markets.

The tenth and eleventh columns of Table 3 show the average values of the variables forfirms with or without bond ratings. While Class 1 is a group of firms without a bond rating,Class 2 is with a bond rating. The fraction of firms with a bond rating is 0.30. The capitalstock and the number of employees of firms without a bond rating are lower. The growthrate of their sales, working capital ratio, and Tobin’s q are higher. The rest of average valuesare close to each other.33See Fazzari, Hubbard, and Petersen (1988).

19

Table 8 exhibits the estimation results for these two groups of firms. The investment rateof firms without a bond rating responds more to changes in the linear term of the fundamen-tals, but less to the squared term, as was the case for expected-to-be financially constrainedfirms identified using other criteria. The investment rate of firms with a bond rating is moresensitive to the cash flow ratio, and to the sales to capital ratio when the mandated invest-ment rate and the profitability shocks are the fundamentals. This indicates that even thoughone would expect that firms with a bond rating should have an easier access to externalfunds, their investment rate is still sensitive to financial variables. But when Tobin’s q isthe fundamental measure, or when the working capital ratio is the financial variable, firmswithout a bond rating respond more to changes in financial variables as expected.

The last two columns of Table 3 report the average values of the variables for firms groupedby the debt to capital ratio. While Class 1 consists of firms with a low debt to capital ratio,Class 2 consists of firms with a high debt ratio. Firms with a low debt to capital ratio have alarger amount of capital on average, but they have a less number of employees. The averageinvestment rate of firms with a high debt to capital ratio is slightly higher. The growth rateof real sales and the earnings retention rate are also higher for these firms.

Table 9 shows the estimation results for these two groups of firms. The sensitivity of theinvestment rate to the ratio of sales to capital and the ratio of working capital to capital ishigher for firms with a high debt to capital ratio. This is an expected result since these firmshave a heavier debt burden; thus, any changes in financial variables should have a largereffect on the investment rate. But the opposite is true when financial variable is representedby the cash flow ratio, and the fundamentals are the profitability shocks and the mandatedinvestment rate. In this case, the investment-cash flow sensitivity is higher for the firms witha low debt ratio. One explanation for this result would be that since firms with a low debtratio may not have enough external funds to finance their investment, any changes in thecash flow ratio should have a larger effect on their investment. But, the investment decisionof firms with a higher debt-to-capital ratio can follow changes in fundamentals more closelysince their debt stock is high, which may indicate that they had already borrowed externalfunds to finance their investment projects.

5 Conclusion

The question investigated in this paper is whether the empirical significance of financialvariables in determining firm-level investment would be explained by measurement errors infundamentals, especially in Tobin’s q. In order to answer this question, it is investigated howthe empirical relationship between the investment rate and financial variables change wheninvestment opportunities are proxied by profitability shocks and the mandated investment

20

rate, both of which would be good alternatives to Tobin’s q. The methodology is the esti-mation of a reduced form investment equation in which different types of fundamentals andfinancial variables are included as explanatory variables.

The results show that financial variables are still statistically significant determinants offirm-level investment even when investment opportunities are proxied by the profitabilityshocks and the mandated investment rate. But the interesting outcome is that the explana-tory power of financial variables for investment drops significantly when the profitabilityshocks and the mandated investment rate are the fundamental variables. For the purpose ofcomparison, the same analyses are repeated using Tobin’s q instead of the recently-introducedfundamentals. The empirical results produced by Tobin’s q are different from the ones men-tioned above, but exactly as expected in the financial market imperfections literature: fi-nancial variables are relatively more significant in explaining investment compared to funda-mentals. These results point out that the statistical and economic significance of financialvariables indeed would be caused by inadequacy of Tobin’s q, or any other fundamental, incapturing investment opportunities.

In addition to full-sample analyses, the link between investment, financial variables, andfundamentals is also investigated in sub-samples which are defined with different a priori cri-teria available in the financial market imperfections literature, used in identifying financiallyconstrained versus unconstrained firms. The sub-sample analyses show that when Tobin’s qis the fundamental variable, the response of investment to changes in financial variables isrelatively higher for financially constrained firms as expected in the literature. On the otherhand, the sensitivity of investment to financial variables, such as the cash flow to capital ra-tio, is lower for expected-to-be financially constrained firms when the profitability shocks andthe mandated investment rate are taken as fundamental determinants of investment. Thisresult is in conflict with the predictions of the finance constraint literature. There might betwo alternative implications of it. On the one hand, assuming that a priori criteria, such asthe level of dividend payout, are sufficient to identify financially constrained firms, we mayconclude that a higher cash flow-investment sensitivity cannot be a good indicator of financialproblems since our results indicate that the sensitivity of investment to internal funds is lowerfor firms taking place in financially constrained groups. This is the issue argued by Kaplanand Zingales (1997) such that the least constrained and the most financially successful firmsin their sample seem to depend primarily on their cash flow to finance their investment de-spite the availability of additional low cost funds. On the other hand, assuming that a highinvestment-cash flow sensitivity is a sufficient indicator of financial problems, our results, inthis case, imply that a priori criteria used in selecting firms with possible financial problemsare not successful in identifying them since investment by firms expected to be financiallyconstrained is relatively less sensitive to internal funds.

21

In terms of future studies related to this topic, the robustness of the results across differentindustries can be investigated. One possibility is that the analyses in this study may focusonly on durable-goods industries, which are more homogenous compared to nondurablesindustries.

22

A Data Appendix

A.1 Sample Selection

The major data source is Standard & Poor’s COMPUSTAT Database, which is a database offinancial, statistical, and market information covering publicly traded companies. The dataset covers only U.S. manufacturing firms with a SIC code between 2000-3999 for the yearsfrom 1981 to 1996. The calendar year, which is a period of one year beginning with January1 and ending with December 31, is taken.

In the initial data set, the number of U.S. manufacturing firms was 3798 active firmsand 3503 inactive firms, for which data are no longer collected for reasons such 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 thecapital expenditure for PPE and for total gross PPE between 1981 and 1996. All foreignfirms 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 ac-tivity. There are different ways to identify these firms. One way is the elimination offirms, which have expenditure on acquisitions accounted for more than 15 percent ofassets. This way has not been used since the data for acquisitions are missing for manyfirms. The other commonly used way is the exclusion of firms whenever

|Gt −Gt−1 − It +Rt| > 0.15 Gt−1,

where Gt is the book value of the capital stock, It is nominal capital expenditure and Rt

is the retirement of capital.34 But, for some firms, the retirement series are missing. Inthis case, the retirement value is set to be zero unless the left hand side of the formula isnegative. In this case, 10 percent of the previous period’s capital is taken as retirementvalue. The idea behind this formula comes from the accounting identity: the bookvalue of physical capital is equal to the capital level of the previous period plus nominalinvestment at the current period minus retirement plus the residual balance, whichincludes changes in capital caused by mergers or acquisitions. So if a firm’s residualbalance is greater than 0.15 of its previous period capital, then it is assumed that thisfirm gets into a merger or acquisition activity, then it is deleted. After applying thismethod, 500 firms are left.

34This method is taken from Gilchrist and Himmelberg (1995).

23

3. Two firms are eliminated since their data are discontinuous due to FASB 94 (FinancialAccounting Standards Board), which requires wholly owned subsidiaries to be includedon the firm’s balance sheet. The total number of firms drops to 498 after deleting thesetwo firms.

4. The investment series is constructed through subtracting the sale of PPE and retirementof PPE series from the capital expenditure on PPE series as defined in Section 3. Dueto the presence of additional missing data points for the sale of capital and retirementseries, the investment rate series could not be constructed for some firms. This factdecreases the number of firms to 471.

5. The replacement value of the capital stock is constructed using a perpetual inventorymethod as specified in Section 3. 8 of firms are deleted since this method producesnegative capital stock for them. After all, the number of firms is 463. The number offirms in each 2-digit industry is presented in Table A1.

A.2 COMPUSTAT series

This section gives information on the definition of the series taken from COMPUSTAT Data-base and their item numbers as reported in COMPUSTAT.

• Assets - Total (item # 6): This item represents current assets plus net property, plant,and equipment plus other noncurrent assets (including intangible assets, deferred items,and investments and advances).

• Capital Expenditures on plant, property and equipment (PPE) (item # 30): This itemrepresents the amount spent for the construction and/or acquisition of property, plant,and equipment.

• Cash and Equivalents (item # 1): It represents cash and all securities readily transfer-able to cash.

• Cash Dividends - Preferred (item # 19): This item represents the total amount of thepreferred dividend requirement on cumulative preferred stock and dividends paid onnoncumulative preferred stock of the company during the year.

• Cash Flow (item # 14 + item # 18): This concept includes Income Before Extra-ordinary Items, which represents the income of a company after all expenses exceptprovisions for common and or preferred dividends, plus Depreciation and Amortiza-tion, which is the non-cash charges for obsolescence of and wear and tear on property.

24

• Common Shares Outstanding - Company (item # 25): This item represents the netnumber of all common shares outstanding at year end, excluding treasury shares andscrip.

• Cost of Goods Sold (item # 41): This item represents all expenses directly allocatedby the company to production, such as material, labor, and overhead.

• Current Assets - Total (item # 4): This item represents cash and other assets whichare expected to be realized in cash or used in the production of revenue within the next12 months.

• Current Liabilities - Total (item # 5): This item represents liabilities due within oneyear, including the current portion of long-term debt.

• Debt - Total (item # 9 + item # 34): This concept is the sum of Total Long-TermDebt, which is defined as debt obligations due more than one year from the company’sbalance sheet date, plus Debt in Current Liabilities, which is defined as the total amountof short-term notes and the current portion of long-term debt (debt due in one year).

• Debt in Current Liabilities (item # 34): This item represents the total amount ofshort-term notes and the current portion of long-term debt (debt due in one year).

• Debt Rating - Current (S&P Senior): This item represents the issuer senior debt ratingthat has been assigned to a company by Standard & Poor’s. An S&P corporate debtrating is a current assessment of the creditworthiness of an obligor with respect to asenior or subordinated debt obligation. This assessment may take into considerationobligors such as guarantors, insurers, or lessees.

• Depreciation and Amortization (item # 14): This item represents non-cash chargesfor obsolescence of and wear and tear on property, allocation of the current portion ofcapitalized expenditures, and depletion charges.

• Dividend Payout (100* (item # 21)/(item # 20): This concept is Cash Dividends-Common, defined as the total dollar amount of dividends (other than stock dividends)declared on the common stock, divided by Income Before Extraordinary Items - Ad-justed For Common Stock Equivalents, which represents income before extraordinaryitems and discontinued operations less preferred dividend requirements (adjusted forcommon stock equivalents). This figure is then multiplied by 100.

• Dividends - Cash Common (item # 21): This item represents the total dollar amount ofdividends (other than stock dividends) declared on the common stock of the companyduring the year.

25

• Dividends - Cash (item # 127): This item represents the total amount of cash dividendsfor both common and preferred stock.

• Earnings Retention Rate (100*(item # 18 - item # 127)/ item # 18): This conceptis Income Before Extraordinary Items minus Cash Dividends. This total is divided byIncome Before Extraordinary Items and then multiplied by 100.

• Employees (item # 29): This item represents the number of company workers as re-ported to shareholders. This is reported by some firms as an average number of em-ployees and by some as the number of employees at year end.. No attempt has beenmade to differentiate between these bases of reporting. If both are given, the year endfigure is used.

• Income Before Extraordinary Items (item # 18): This item represents the incomeof a company after all expenses, including special items, income taxes, and minorityinterest - but before provisions for common and/or preferred dividends. This item doesnot reflect discontinued operations (appearing below taxes) or extraordinary items.

• Income Before Extraordinary Items - Adjust.. for Common Stock Equivalents (item# 20): This item represents income before extraordinary items and discontinued op-erations less preferred dividend requirements and is adjusted for the additional dollarsavings due to common stock equivalents.

• Interest Expense (item # 15): This item represents the periodic expense to the companyof securing short- and long-term debt.

• Inventories - Finished Goods (item # 78): This item represents manufactured goodsready for sale. Merchandise in transit to customers is included.

• Inventories - Total (item # 3): This item represents merchandise bought for resale andmaterials and supplies purchased for use in production of revenue.

• Long-Term Debt - Total (item # 9): The item represents debt obligations due morethan one year from the company’s balance sheet date.

• Net Income (Loss) (item # 172): This item represents the income or loss reported bya company after expenses and losses have been subtracted from all revenues and gainsfor the fiscal period including extraordinary items and discontinued operations.

• Operating Income Before Depreciation (item # 13): This item represents Sales (Net)less Cost of Goods Sold and Selling, General, and Administrative Expense before de-ducting Depreciation, Depletion, Amortization, interest expense, and taxes.

26

• Price - Calendar Year - Close (item # 24): These items represent the absolute high,low and close transactions during the year for companies on national stock exchangesand bid prices for over-the-counter issues.

• Plant, Property, and Equipment (PPE) total gross (item # 7): This item representsthe cost of tangible fixed property used in the production of revenue.

• PPE total net (item # 8): This item represents the cost, less accumulated depreciation,of tangible fixed property used in the production of revenue.

• Retained Earnings (item # 36): This item represents the cumulative earnings of thecompany less total dividend distributions to shareholders.

• Retirements (PPE) (item # 184): This item represents a deduction from a company’sproperty, plant, and equipment account resulting from the retirement of obsolete ordamaged goods and/or physical structures.

• Sale of Property, Plant, and Equipment (item # 107): This item represents fundsreceived or cash in flows from the sale of property, plant, and equipment.

• Sales (Net) (item # 12): This item represents gross sales (the amount of actual billingsto customers for regular sales completed during the period) reduced by cash discounts,trade discounts, and returned sales and allowances for which credit is given to customers.

• Short-Term Borrowings (average) (item # 104): This item represents the approximateaggregate short-term financing outstanding during the company’s reporting year. Short-term borrowings are usually in the form of credit with banks.

• Standard Industrial Classification (SCI) code: This filed contains a four-digit SIC num-ber.

• Tax Rate: This item is Total Income Taxes, which include income taxes imposed byfederal, state, and foreign governments, divided by Pretax Income, which is operatingand nonoperating income before provisions for income taxes and minority interest. Thisis then multiplied by 100.

• Working Capital (item # 179): This item represents the difference between total currentassets minus total current liabilities as reported on a company’s Balance Sheet.

27

B Details on the Calculation of Fundamentals

B.1 Tobin’s q

The numerator of average Tobin’s q is the sum of the market value of common stock, theliquidating value of preferred stock, the market value of long-term debt, and the book valueof short-term debt. The denominator is the sum of replacement value of fixed capital andinventories. These variables are calculated as follows:

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 forthem. For firms using the last-in-first-out (LIFO) or any other method, inventories are valuedat 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:

INVt = INVt−1(PPItPPIt−1

) + INV ∗t − INV ∗t−1 if INV ∗t ≥ INV ∗t−1

INVt = (INVt−1 + INV ∗t − INV ∗t−1) (PPItPPIt−1

) if INV ∗t < INV ∗t−1

where INVt is the replacement value of LIFO inventories at time t and INV ∗t is their reportedbook value.

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 thereplacement value. Since the COMPUSTAT database provides only limited information onmaturities of debt, it is necessary to construct the maturity distribution of long-term debtfrom historical information on debt issues. Firstly, following Brainard, Shoven and Weiss(1980), it is assumed that all long-term debts mature in twenty years. For the first year, eachindividual firm’s maturity distribution is set equal to the aggregate taken from HistoricalStatistics of the United States, series X 499-509, p. 1005 for the years 1961-1970. I give equalweight to the maturity distribution for years 1971-80. Then, if Djt is debt due in j years attime t, LTDt is the reported value of long-term debt at time t, and DIt is the amount ofdebt 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

andDjt = Dj+1,t−1, j = 1, ..., 19.

28

If LTDt − (LTDt−1 −D1,t−1) < 0, debt due in one to nineteen years is scaled down bythe factor:

LTDt

LTDt−1 −D1,t−1.

The actual values of debt due in one to five year are available in COMPUSTAT. Thisvalues are replaced by the calculated values and the rest of the maturity distribution isrescaled in order to be consistent with total amount of long-term debt:

Dajt = D∗jt, j = 1, ..., 5;

Dajt = Djt

⎛⎜⎜⎜⎝1 +5P

j=1(Djt −D∗jt)

20Pj=6

Djt

⎞⎟⎟⎟⎠ , j = 6, ..., 20;

where Dajt 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.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 attime t is the Baa rate at time t. The new value of total book value is scaled as follows:

NLTDt = LTDt ×IEXt

20Pj=1

Baat+j−20 Dajt

where NLTDt is the scaled value of long-term debt at time t, Baat is the interest rate ongrade Baa bonds at time t and IEXt is the book value of interest expense at time t. Thenthe new maturity distribution is set proportional to the old distribution.

Market value of Equity: The value of common stock at the beginning of each year isestimated, following Salinger and Summers (1983), as the closing price of a share of stock foreach company in year t − 1 times the number of outstanding shares at t − 1. The value ofpreferred stock is estimated by dividing preferred cash dividends by the Standard and Poor’spreferred stock yield (taken from CITIBAS database).

B.2 Frictionless Capital

The method to obtain the frictionless capital stock is taken from Caballero, Engel, andHaltiwanger (1995), and applied to the firm-level data.35 All frictions are assumed to beabsent, including time-to-build assumption, and any adjustment costs. Y represents the35See Bayraktar (2002) for details.

29

value of output of an individual firm, where imperfect competition is assumed, and fixedfactors other than capital produce a decreasing returns:

Y = AKαKLαL , αK + αL < 1,

where A,K, and L are profitability shocks, capital stock, and flexible factors, respectively.αK is the capital share of production and αL is the one for the flexible factor.

Optimizing over flexible factors yields a profit function:

Π(A,K) ≡ maxL

Y − wLL,

where wL is the price of flexible factors. Given this equation, frictionless capital is defined as

K∗ = argmaxKΠ(A,K)− cK.

where c is the cost of capital.After some manipulations and taking the logarithm of the above expression, the friction-

less capital level is expressed as

k∗ − k = η{y − k − c},

where η ≡ (1− αL)/(1− αK − αL), and y is equal to the log of real output, which is definedas the sum of the real sale value of goods plus changes in the real value of finished goodsinventories, and k is equal to the log of real capital. η is a decreasing function of the curvatureof the profit function with respect to capital, which is approximately equal to:36

η ≈1

1− α,

where α is the cost share of capital, which is estimated at the 4-digit industry level. The costof capital, c, is defined as

(rt + δt)pitpt

ϕt

where rt is the real interest rate, which is equal to the average nominal Baa corporate bondrate minus the measure of expected inflation from the Livingston Survey of twelve-monthinflation expectations. δt is the depreciation rate taken from the 2-digit unofficial BLS dataset. In this set, depreciation rates are given for three asset groups for each 2-digit industry. Iused the wealth share of assets as a weight to calculate the average depreciation rate in each2-digit industry group. pit is the new capital expenditures deflator and pt is shipments price

36Details on the calibration of parameters are given in Table A2.

30

deflator taken from Gray and Bartelsman data set (at the 4-digit industry level). The taxparameter, ϕt, is taken from the BLS database such that:

ϕt =(1− τ tzt − κt)

1− τ t

where τ t is the corporate income tax, zt is the present value of $1 of tax depreciation al-lowances, and κt is the effective rate of investment tax credit. This equation is given for93 assets for each two-digit industry level. In order to find the two-digit weighted averagevalues, the wealth share of assets in each industry is used.

B.3 Calculation of profitability shocks

This section explains the second way of calculating profitability shocks through the first ordercondition of maximizing a profit function with respect to labor. It is assumed that a firmmaximizes the following profit function with respect to labor:

Π(A,K) = maxL

R( bA,K,L)− Lw

where A is the profitability shock that contains both aggregate and idiosyncratic shocks, bAis the shock to the revenue function, K is the capital stock, L is labor, and w represents thewage of labor. Assuming that the product market is imperfectly competitive, the revenuefunction is defined as

R( bA,K,L) = bApy = bAyξy = bAy1+ξ,where the constant returns to scale Cobb-Douglas production function is assumed:

y = KαKLαL ,

where αL and αK are the production function coefficients on labor and capital. The demandcurve is given as

p = yξ,

where ξ is the elasticity of the demand curve. Thus, the revenue function equals:

R( bA,K,L) = bAyξy = bAy1+ξ.From the first order condition with respect to L, the optimum value of L is:

L∗ =

" bA(1 + ξ)αLw

#− 1αL(1+ξ)

KαK (1+ξ)

1−αL(1+ξ) .

31

After plugging this optimum labor back into the profit equation, it becomes:

Π(A,K) = (A1 −A2)Kθ,

where θ = (1−αL)(1+ξ)1−αL(1+ξ) , A1 =

bA h bA(1+ξ)αLw

i αL(1+ξ)

1−αL(1+ξ) , A2 = wh bA(1+ξ)αL

w

i 11−αL(1+ξ) , and from

L∗ equation bA is defined as

bA = w

(1 + ξ)αL

∙L

Kθ

¸1−αL(1+ξ).

Note that (A1 −A2) is the profitability shock, A, which is equal to

A = bA 11−αL(1+ξ)w

−αL(1+ξ)1−αL(1+ξ)

½[(1 + ξ)αL]

αL(1+ξ)

1−αL(1+ξ) − [(1 + ξ)αL]1

1−αL(1+ξ)

¾.

In terms of calibration, θ is estimated by regressing the log of real profit data on the logof real capital using firm-level panel data. Its value is estimated at 0.61. αL is estimatedas 0.73 again using firm-level data.37 ξ is obtained from θ = (1−αL)(1+ξ)

1−αL(1+ξ) equation. These

coefficients imply a demand elasticity of -0.15 and a markup of about 18 percent.38

37αL is estimated using cost shares. The cost share of labor is calculated as the ratio of wages timesemployment level to the sum of the rental price of capital times capital level and wages times employmentlevel. The employment and capital data are from COMPUSTAT database. While the rental price of capitalseries is obtained from BLS database, the wage data is from Gray and Bartelsman’s 4-digit dataset.38Details on the calibrated parameters are given in Table A3.

32

References

[1] Abel, Andrew B., 1979, Investment and the Value of Capital, New York: Garland.

[2] Abel, Andrew B.; and Janice C. Eberly, 2002, "Investment and q with Fixed Costs: anEmpirical Analysis," mimeo, University of Pennsylvania (January).

[3] Abel, Andrew B.; and Janice C. Eberly, 2003, "Q Theory Without Adjustment Costs andCash Flow Effects Without Financing Constraints," mimeo, University of Pennsylvania(October).

[4] Barnett, Steven; and Plutarchos Sakellaris, 1998, "Nonlinear Response of Firm Invest-ment to Q: Testing a Model of Convex and Non-convex Adjustment Costs," Journal ofMonetary Economics, vol. 42, pp: 261-288.

[5] Barnett, Steven; and Plutarchos Sakellaris, 1999, "A New Look at Firm Market Value,Investment and Adjustment Cost," The Review of Economics and Statistics, vol. 81, pp:250-60.

[6] Bayraktar, Nihal, 2002, "Analyses of Alternative Fundamentals of Fixed Capital Invest-ment," University of Maryland, mimeo.

[7] Bayraktar, Nihal; Plutarchos Sakellaris; and Philip Vermeulen, 2005, "Real versus Fi-nancial Frictions to Capital Investment," mimeo.

[8] Bernanke, Ben S.; and John Y. Campbell, 1988, "Is there a Corporate Debt Crisis?"Brookings Papers on Economic Activity, vol.1, pp.83-139.

[9] Bernanke, Ben S.; John Y. Campbell; and Toni Whited, 1990, "US Corporate Leverage:Developments in 1987 and 1988," Brookings Papers on Economic Activity, no.1, pp:255-286.

[10] Bernanke, Ben; Mark Gertler, 1990, "Financial Fragility and Economic Performance,"Quarterly Journal of Economics, pp: 87-114 (February).

[11] Bertola, Guiseppe; and Ricardo J. Caballero, 1994, "Irreversibility and Aggregate In-vestment," Review of Economic Studies, vol. 61, pp. 223-246 (April).

[12] Brainard William C.; John B. Shoven and Laurence Weiss, 1980, "The Financial Valu-ation of the Return to Capital," Brookings Papers on Economic Activity, vol.2, pp-453-511.

33

[13] Caballero, Ricardo J. and Eduardo M.R.A. Engel, 1994, "Explaining Investment Dy-namics in the U.S. Manufacturing : A Generalized (S,s) Approach," NBER WorkingPaper No. 4887.

[14] Caballero, Ricardo J., Eduardo M.R.A. Engel, and John Haltiwanger, 1995, "Plant-Level Adjustment and Aggregate Investment Dynamics," Brookings Papers on EconomicActivity, vol 2, pp.1-39.

[15] Calomiris, Charles W.; Charles P. Himmelberg; and Paul Wachtel, 1994, "CommercialPaper, Corporate Finance , and The Business Cycle: A Microeconomic Perspective,"NBER Working Paper, no. 4848 (September).

[16] Carpenter R. E.; S. M. Fazzari; and B.C. Petersen, 1998, "Financing Constraints andInventory Investment: a Comparative Study with High-Frequency Panel Data", Reviewof Economics and Statistics, vol. 80, pages 513-519.

[17] Cooper, Russell W.; and Joao Ejarque, 2003, "Exhuming Q: Market Power or CapitalMarket Imperfections," mimeo, University of Texas at Austin (November).

[18] Cooper, Russell W.and John C. Haltiwanger, 2002, "On the Nature of Capital Adjust-ment Costs", unpublished (January).

[19] Cooper, Russell W.and John C. Haltiwanger, 2005, "On the Nature of Capital Adjust-ment Costs", unpublished (January).

[20] Erickson, Timothy; and Toni M. Whited, 2000, "Measurement Error and the Rela-tionship between Investment and q," Journal of Political Economy, vol. 108(5), pages1027-57.

[21] Fazzari, Steven M.; Glenn R. Hubbard; and Glenn and Bruce C. Petersen, 1988, "Financ-ing Constraints and Corporate Investment," Brookings Paper for Economic Activity, no.1, pp. 141-195.

[22] Gilchrist, Simon; and Charles P. Himmelberg, 1995, "Evidence on the Role of Cash Flowfor Investment," Journal of Monetary Economics, vol. 36, pp. 541-572 (December).

[23] Gilchrist, Simon; and Charles P. Himmelberg, 1998, "Investment, Fundamentals andFinance,” NBER Working Paper # 6652 (July).

[24] Gomes, Joao F., 2001, "Financing Investment," American Economic Review, vol. 91, vo:5, pp. 1263-85 (December).

34

[25] Goolsbee Austan; and David B. Gross, 1997, "Estimating Adjustment Costs with Dataon Heterogeneous Capital Goods," NBER WP No. 6342 (December).

[26] Hayashi, F., 1982, "Tobin’s Marginal Q and Average Q: A Neoclassical Interpretation,"Econometrica, pp:213-224 (January).

[27] Hoshi, Takeo; Anil Kashyap; and David Scharfstein, 1991, "Corporate Structure, Liq-uidity, and Investment: Evidence from Japanese Industrial Groups," Quarterly Journalof Economics, vol. 106, no.1, pp. 33-60 (February).

[28] Hu, Xiaoqiang; and Fabio Schiantarelli, 1998, "Investment and Capital Market Imper-fections: A Switching Regression Approach Using U.S. Firm Panel Data," The Reviewof Economics and Statistics, vol.80, no.3, pp: 466-479 (August).

[29] Hubbard, R. Glenn, 1998, "Capital-Market Imperfections and Investment," Journal ofEconomic Literature, vol. 36, pp. 193-225 (March).

[30] Hubbard, R. Glenn; Anil K. Kashyap; and Toni M. Whited, 1995, "Internal Financeand Firm Investment," Journal of Money, Credit and Banking, vol.27, no.3, pp.683-701(August).

[31] Jaramillo, Fidel; Fabio Schiantarelli; and Andrew Weiss, 1996, "Capital Market Imper-fections Before and after Financial Liberalization: An Euler Equation Approach to Paneldata for Ecuadorian Firms," Journal of Development Economics, vol.51, pp.367-386.

[32] Kaplan Steven N.; and Luigi Zingales, 1995, "Do Financing Constraints Explain WhyInvestment is Correlated with Cash Flow?" NBERWorking Paper. no. 5267 (September).

[33] Kaplan Steven N.; and Luigi Zingales, 1997, "Do Investment Cash Flow Sensitive ProvideUseful Measure of Financing Constraints?” Quarterly Journal of Economics, vol. 112,no. 1, pp. 169-215 (February).

[34] Kaplan Steven N.; and Luigi Zingales, 2000, "Investment-Cash Flow Sensitivities arenot Valid Measures of Financing Constraints," NBER Working Paper, no. 7659 (April).

[35] Kashyap, Anil K.; Owen A. Lamont; and Jeremy Stein, 1994, "Credit Conditions andthe Cyclical Behavior of Inventories," Quarterly Journal of Economics, vol. 109(3), pages565-92 (August).

[36] Salinger Micheal and Lawrence H. Summers, 1983, "Tax Reform and Corporate Invest-ment: A Microeconomic Simulation Study," Feldstein Martin (ed) Behavioral SimulationMethods in Tax Policy Analysis.

35

[37] Schiantarelli, Fabio, 1996, "Financial Constraints and Investment: Methodological Issuesand International Evidence," Oxford Review of Economic Policy, vol. 12(2), pages 70-89(Summer).

[38] Whited, Toni M., 1992, "Debt, Liquidity Constraints, and Corporate Investment: Evi-dence from Panel Data," Journal of Finance, vol. 47, no. 4, pp. 1425-1460 (September).

36

Figure 1 Investment Rate Distribution, 1983-1996

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

-0.10

-0.05 0.0

00.0

50.1

00.1

50.2

00.2

50.3

00.3

50.4

00.4

50.5

0

Investment rate

Frac

tion

of fi

rms

Table 1 Summary Statistics, 1983-1996

Variable MeanStandard deviation Min

25th percentile

50th percentile

75th percentile Max

Plant, Property, and Equipment (PPE) (in millions of 1992 dollars) 1511.12 6943.51 0.09 18.71 79.28 397.74 100263.35PPE (gross book value) (in millions of dollars) 1880.46 8121.73 0.09 22.50 97.90 533.00 126000.00Investment rate 0.13 0.34 -0.72 0.05 0.10 0.16 22.20Employees (in thousands) 13.40 48.95 0.00 0.49 1.93 7.70 876.80Debt burden (flow) 0.04 0.08 0.00 0.01 0.03 0.06 4.50Debt burden (stock) 197.21 6528.13 0.00 0.04 0.16 0.50 375146.26Total Debt over PPE 0.39 0.65 0.00 0.12 0.28 0.49 27.20Interest expenditure coverage rat -0.03 12.98 -916.00 0.04 0.12 0.22 244.00Dividends over PPE 0.04 0.18 0.00 0.00 0.02 0.05 4.71Dividends payout ratio 0.30 4.11 -213.93 0.00 0.19 0.41 152.56Growth rate of real (net) sales 0.15 4.24 -0.99 -0.02 0.05 0.15 338.00Earnings retention rate 0.68 3.86 -151.56 0.59 0.80 1.00 214.93Cash flow to PPE ratio 0.21 0.49 -21.10 0.10 0.18 0.28 8.48Sales (net) to PPE ratio 2.86 2.80 0.00 1.54 2.24 3.33 59.70Working capital to PPE ratio 0.86 1.89 -2.77 0.26 0.53 0.98 76.70Cash and equivalence to PPE rat 0.35 1.36 0.00 0.03 0.10 0.30 64.00Tax rate 31.92 203.75 -7840.00 29.60 36.90 41.40 7340.00

FundamentalsProfitability shocks 0.00 0.20 -0.69 -0.12 0.00 0.11 0.70Profitability shocks (residuals) 0.01 0.28 -1.13 -0.13 0.01 0.15 0.94Mandated inv. rate 0.02 0.29 -1.13 -0.14 0.02 0.16 1.14Log of Tobin's q 0.25 0.50 -0.94 -0.06 0.19 0.50 2.48

Table 2 Response of Investment to Alternative Measures of Fundamentals and Financial

Variables (Full Sample Results)

Financial Variables None CF_K Sales_K WorkingK_K(1) (2) (3) (4)

Profitability shocks (first-order)

ait 0.229 0.22 0.179 0.202(22.380) (21.745) (16.880) 19.759

ait2 0.244 0.225 0.224 0.232(6.626) (6.211) (6.202) 6.398

Financial variable 0.035 0.033 0.055(4.247) (14.525) (15.422)

Adj. R2 0.227 0.233 0.257 0.261

Profitability shocks (residuals)

rait 0.125 0.118 0.066 0.099(16.103) (14.325) (7.855) (12.567)

rait2 0.108 0.092 0.072 0.082(6.168) (5.357) (4.214) (4.728)

Financial variable 0.042 0.040 0.054(3.793) (16.551) (14.322)

Adj. R2 0.199 0.208 0.239 0.229

Mandated inv. rate

kit 0.196 0.193 0.180 0.173(26.963) (25.456) (19.529) (22.204)

kit2 0.071 0.071 0.069 0.066(4.954) (4.987) (4.752) (4.562)

Financial variable 0.002 0.008 0.033(0.209) (2.802) (8.093)

Adj. R2 0.252 0.257 0.253 0.261

Tobin's q

qit 0.197 0.162 0.100 0.131(7.379) (6.369) (4.422) (5.126)

qit2 0.043 0.031 0.031 0.033(2.234) (1.702) (1.936) (1.801)

Financial variable 0.478 0.166 0.208(20.837) (44.058) (26.169)

Adj. R2 0.056 0.134 0.325 0.173

Note: The estimation technique is OLS. Time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variable is the investment rate. ait is the first-order profitability shocks, rait is the profitability shocks (residuals), kit is the mandated investment rate, and qit is Tobin's q. CF_K stands for the cashflow-to-capital ratio, Sales_K stands for the net sales to capital ratio, WorkingK_K is the ratio of working capital to capital.

Firm

s with

sm

all

capi

tal

stoc

k

Firm

s with

la

rge

capi

tal

stoc

k

Firm

s with

lo

w

num

ber o

f em

ploy

ees

Firm

s with

hi

gh

num

ber o

f em

ploy

ees

Firm

s with

lo

w

divi

dend

pa

yout

ra

tio

Firm

s wi

th

high

di

vide

nd

payo

ut

ratio

Firm

s with

lo

w

divi

dend

to

capi

tal

ratio

Firm

s with

hi

gh

divi

dend

to

capi

tal

ratio

Firm

s w

ithou

t bo

nd r

atin

gFi

rms

with

bo

nd ra

ting

Firm

s wi

th

low

deb

t to

capi

tal

ratio

Firm

s with

hi

gh d

ebt

to c

apita

l ra

tio

Plan

t, Pr

oper

ty, a

nd

Equi

pmen

t (PP

E, in

mill

ions

of

1992

dol

lars

)27

.71

3034

.49

43.0

430

05.5

096

5.70

2057

.19

437.

2225

92.7

523

2.01

4805

.24

1823

.18

1214

.08

PPE

(gro

ss b

ook

valu

e, i

n m

illio

ns o

f dol

lars

)33

.05

3322

.61

51.2

332

89.4

810

84.3

622

28.9

849

8.36

2822

.17

256.

0052

61.3

419

22.4

314

06.0

9In

vest

men

t rat

e0.

130.

120.

140.

120.

150.

110.

140.

110.

130.

120.

120.

14Em

ploy

ees (

in th

ousa

nds)

0.86

25.2

40.

6625

.25

11.2

814

.58

4.61

21.2

13.

0838

.07

9.80

16.0

7G

row

th ra

te o

f rea

l (ne

t) sa

les

0.21

0.07

0.22

0.07

0.24

0.05

0.23

0.06

0.17

0.08

0.08

0.21

Earn

ings

rete

ntio

n ra

te0.

810.

560.

900.

471.

040.

340.

910.

470.

690.

670.

610.

76To

tal D

ebt o

ver P

PE0.

410.

350.

410.

350.

460.

300.

480.

290.

380.

390.

160.

60D

ivid

ends

pay

out r

atio

0.22

0.40

0.11

0.51

-0.0

90.

700.

080.

540.

320.

300.

390.

23D

ivid

ends

ove

r PPE

0.05

0.04

0.04

0.05

0.01

0.08

0.01

0.08

0.05

0.04

0.06

0.03

Cas

h flo

w to

PPE

ratio

0.22

0.20

0.21

0.21

0.18

0.24

0.15

0.27

0.22

0.20

0.26

0.16

Sale

s (ne

t) to

PPE

ratio

3.25

2.33

3.02

2.57

2.82

2.77

2.64

2.94

3.01

2.25

2.56

3.03

Wor

king

cap

ital t

o PP

E ra

tio1.

130.

471.

070.

530.

910.

710.

810.

810.

970.

380.

910.

70

Fund

amen

tals

Prof

itabi

lity

shoc

ks (f

irst-o

rder

)0.

000.

000.

000.

000.

000.

000.

000.

000.

000.

000.

000.

00Pr

ofita

bilit

y sh

ocks

(res

idua

ls)0.

010.

000.

010.

000.

010.

000.

010.

000.

010.

000.

010.

01M

anda

ted

inv.

rate

0.

020.

010.

010.

020.

020.

010.

020.

010.

020.

010.

020.

01Lo

g of

Tob

in's

q 0.

370.

130.

370.

140.

360.

150.

280.

220.

300.

120.

240.

26

Div

iden

d to

cap

ital

ratio

Bon

d ra

ting

Tot

al d

ebt t

o ca

pita

l ra

tioSi

ze o

f cap

ital s

tock

Num

ber

of e

mpl

oyee

sD

ivid

end

payo

ut r

atio

Tab

le 3

A

vera

ge V

alue

s of V

aria

bles

Whe

n Fi

rms a

re C

lass

ified

ac

cord

ing

to D

iffer

ent F

irm

Cha

ract

eris

tics

Table 4 Response of Investment to Alternative Measures of Fundamentals and Financial

Variables (When Firms are grouped by the size of their capital stock)

Financial Variables

Class 1 Class 2 Class 1 Class 2 Class 1 Class 2 Class 1 Class 2(1) (2) (3) (4) (5) (6) (7) (8)

Profitability Shocks (first-order)

ait 0.260 0.155 0.241 0.148 0.216 0.109 0.231 0.131(17.634) (12.294) (16.582) (11.844) (13.827) (8.588) (15.661) (10.259)

ait2 0.198 0.263 0.167 0.271 0.185 0.247 0.190 0.234(3.880) (5.289) (3.372) (5.508) (3.655) (5.148) (3.783) (4.752)

Financial variable 0.064 0.138 0.022 0.058 0.043 0.125(5.502) (7.677) (7.922) (13.687) (10.196) (11.208)

Adj. R2 0.221 0.256 0.233 0.272 0.239 0.306 0.251 0.291

rait 0.130 0.116 0.127 0.100 0.080 0.051 0.100 0.099(11.907) (11.073) (10.888) (9.203) (6.685) (4.567) (9.023) (9.340)

rait2 0.105 0.139 0.086 0.130 0.074 0.101 0.077 0.112(4.430) (5.336) (3.729) (5.021) (3.158) (3.975) (3.286) (4.292)

Financial variable 0.032 0.100 0.029 0.063 0.048 0.132(2.257) (5.130) (9.550) (13.563) (10.735) (11.543)

Adj. R2 0.181 0.237 0.194 0.244 0.209 0.287 0.216 0.275

kit 0.200 0.185 0.202 0.176 0.195 0.142 0.174 0.164(18.820) (19.138) (18.020) (17.731) (14.139) (11.403) (14.999) (16.000)

kit2 0.083 0.050 0.082 0.054 0.081 0.052 0.076 0.040(3.953) (2.610) (4.000) (2.815) (3.850) (2.719) (3.631) (2.058)

Financial variable-0.017 0.070 0.002 0.029 0.029 0.083

(-1.328) (3.823) (0.681) (5.269) (5.615) (7.223)

Adj. R2 0.223 0.313 0.229 0.316 0.223 0.320 0.233 0.326

Tobin's q

qit 0.282 0.105 0.236 0.095 0.130 0.083 0.165 0.091(5.408) (8.493) (4.762) (7.789) (3.001) (6.982) (3.361) (7.258)

qit2 0.010 0.037 0.005 0.027 0.019 0.029 0.014 0.036(0.329) (2.456) (0.153) (1.798) (0.740) (2.057) (0.481) (2.410)

Financial variable0.535 0.151 0.181 0.057 0.214 0.106

(15.575) (7.764) (32.885) (15.253) (18.670) (10.036)

Adj. R2 0.047 0.263 0.135 0.280 0.344 0.326 0.168 0.294

Class 1: small capital stock Class 2: large capital stock None

Profitability Shocks (residuals)

Mandated inv. Rate

WorkingK_KSales_KCF_K

Note: The estimation technique is OLS. 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), kit is the mandated investment rate, and qit is Tobin's q. 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. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by their average capital stock size.

Table 5 Response of Investment to Alternative Measures of Fundamentals and Financial

Variables (When Firms are grouped by the number of employees)

Financial Variables

Class 1 Class 2 Class 1 Class 2 Class 1 Class 2 Class 1 Class 2(1) (2) (3) (4) (5) (6) (7) (8)

Profitability Shocks (first-order)

ait 0.256 0.165 0.239 0.149 0.213 0.116 0.227 0.140(17.145) (13.195) (16.302) (12.106) (13.563) (9.189) (15.176) (11.237)

ait2 0.177 0.304 0.146 0.317 0.163 0.291 0.165 0.288(3.432) (6.173) (2.912) (6.593) (3.195) (6.107) (3.271) (5.954)

Financial variable 0.054 0.210 0.023 0.048 0.042 0.129(4.670) (11.164) (7.687) (13.461) (9.706) (12.762)

Adj. R2 0.224 0.246 0.234 0.280 0.241 0.295 0.251 0.291

rait 0.126 0.126 0.127 0.093 0.078 0.059 0.097 0.103(11.462) (12.028) (10.954) (8.479) (6.575) (5.242) (8.743) (9.695)

rait2 0.104 0.148 0.087 0.135 0.074 0.112 0.077 0.123(4.430) (5.522) (3.784) (5.101) (3.173) (4.301) (3.308) (4.605)

Financial variable 0.019 0.175 0.030 0.052 0.048 0.123(1.373) (8.467) (9.399) (12.951) (10.476) (11.852)

Adj. R2 0.186 0.225 0.198 0.245 0.214 0.271 0.220 0.265

kit 0.199 0.187 0.201 0.168 0.191 0.153 0.173 0.165(18.788) (19.149) (18.300) (16.445) (14.143) (12.378) (15.100) (15.952)

kit2 0.071 0.069 0.070 0.070 0.070 0.064 0.065 0.058(3.452) (3.458) (3.463) (3.511) (3.373) (3.203) (3.173) (2.886)

Financial variable-0.021 0.111 0.003 0.020 0.029 0.081

(-1.630) (5.664) (0.907) (4.446) (5.502) (7.713)

Adj. R2 0.228 0.305 0.234 0.313 0.228 0.310 0.237 0.320

Tobin's q

qit 0.303 0.093 0.251 0.079 0.132 0.076 0.179 0.079(5.717) (7.754) (4.950) (6.819) (2.992) (6.603) (3.588) (6.542)

qit2 0.001 0.046 0.000 0.018 0.016 0.035 0.009 0.040(0.018) (3.036) (0.011) (1.245) (0.614) (2.464) (0.307) (2.660)

Financial variable0.491 0.329 0.186 0.054 0.212 0.125

(14.563) (14.500) (33.158) (16.186) (18.402) (11.932)

Adj. R2 0.050 0.238 0.128 0.297 0.351 0.311 0.169 0.282

Profitability Shocks (residuals)

Mandated inv. Rate

Class 1: low number of employees Class 2: high number of employeesNone CF_K Sales_K WorkingK_K

Note: The estimation technique is OLS. 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), kit is the mandated investment rate, and qit is Tobin's q. 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. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the number of their employees.

Table 6 Response of Investment to Alternative Measures of Fundamentals and Financial

Variables (When Firms are grouped by the dividend payout ratio)

Financial Variables

Class 1 Class 2 Class 1 Class 2 Class 1 Class 2 Class 1 Class 2(1) (2) (3) (4) (5) (6) (7) (8)

Profitability Shocks (first-order)

ait 0.244 0.189 0.226 0.180 0.186 0.165 0.204 0.186(15.383) (18.110) (14.403) (17.490) (10.900) (15.901) (12.760) (17.664)

ait2 0.171 0.320 0.144 0.316 0.157 0.304 0.160 0.313(3.141) (7.643) (2.706) (7.686) (2.928) (7.434) (3.007) (7.439)

Financial variable 0.067 0.121 0.031 0.025 0.057 0.023(5.186) (9.406) (8.488) (11.328) (10.936) (5.067)

Adj. R2 0.220 0.261 0.232 0.285 0.241 0.295 0.256 0.268

rait 0.130 0.133 0.127 0.113 0.069 0.099 0.096 0.129(11.024) (15.617) (10.224) (12.244) (5.283) (10.829) (8.021) (14.531)

rait2 0.111 0.120 0.092 0.116 0.066 0.110 0.070 0.121(4.332) (5.702) (3.654) (5.546) (2.591) (5.309) (2.787) (5.709)

Financial variable 0.034 0.079 0.040 0.022 0.065 0.015(2.207) (5.491) (10.465) (9.330) (11.859) (3.193)

Adj. R2 0.191 0.226 0.202 0.235 0.225 0.250 0.234 0.230

kit 0.212 0.168 0.210 0.162 0.201 0.153 0.176 0.169(18.915) (20.411) (18.163) (18.037) (13.428) (15.596) (14.469) (19.417)

kit2 0.048 0.137 0.047 0.135 0.046 0.131 0.042 0.140(2.289) (6.915) (2.294) (6.804) (2.218) (6.627) (2.041) (6.985)

Financial variable-0.005 0.021 0.005 0.007 0.047 -0.003

(-0.370) (1.477) (1.097) (2.742) (7.370) (-0.710)

Adj. R2 0.246 0.290 0.252 0.291 0.246 0.292 0.262 0.289

Tobin's q

qit 0.350 0.057 0.283 0.053 0.123 0.051 0.181 0.056(6.476) (5.207) (5.516) (4.898) (2.842) (4.824) (3.576) (5.032)

qit2 -0.014 0.040 -0.011 0.032 0.023 0.030 0.017 0.035(-0.413) (3.424) (-0.345) (2.764) (0.850) (2.664) (0.556) (2.919)

Financial variable0.590 0.126 0.219 0.030 0.271 0.021

(16.095) (8.961) (37.346) (12.749) (21.223) (4.267)

Adj. R2 0.055 0.174 0.149 0.200 0.408 0.224 0.208 0.181

Profitability Shocks (residuals)

Mandated inv. Rate

Class 1: low dividend payout ratio Class 2: high dividend payout ratioNone CF_K Sales_K WorkingK_K

Note: The estimation technique is OLS. 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), kit is the mandated investment rate, and qit is Tobin's q. 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. 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 dividend payout ratio.

Table 7 Response of Investment to Alternative Measures of Fundamentals and Financial

Variables (When Firms are grouped by the dividend to capital ratio)

Financial Variables

Class 1 Class 2 Class 1 Class 2 Class 1 Class 2 Class 1 Class 2(1) (2) (3) (4) (5) (6) (7) (8)

Profitability Shocks (first order)

ait 0.227 0.214 0.212 0.195 0.175 0.179 0.192 0.203(14.585) (19.556) (13.874) (18.164) (10.560) (16.419) (12.307) (18.605)

ait2 0.129 0.412 0.108 0.384 0.125 0.377 0.133 0.381(2.461) (9.006) (2.112) (8.605) (2.404) (8.490) (2.567) (8.393)

Financial variable 0.057 0.140 0.030 0.027 0.058 0.038(4.274) (12.140) (8.069) (13.093) (10.088) (10.352)

Adj. R2 0.213 0.294 0.223 0.331 0.233 0.337 0.244 0.322

rait 0.125 0.138 0.128 0.101 0.072 0.088 0.096 0.121(10.822) (15.488) (10.581) (10.402) (5.747) (9.137) (8.199) (13.131)

rait2 0.117 0.116 0.099 0.106 0.074 0.100 0.078 0.109(4.620) (5.287) (3.995) (4.922) (2.951) (4.663) (3.139) (4.952)

Financial variable 0.018 0.118 0.038 0.028 0.065 0.035(1.104) (8.811) (9.699) (11.759) (10.895) (9.065)

Adj. R2 0.193 0.238 0.203 0.259 0.222 0.276 0.229 0.261

kit 0.211 0.164 0.208 0.156 0.199 0.144 0.175 0.164(19.127) (19.391) (18.476) (16.442) (13.675) (13.939) (14.775) (18.024)

kit2 0.063 0.086 0.061 0.084 0.060 0.083 0.057 0.087(2.987) (4.626) (2.994) (4.517) (2.870) (4.445) (2.732) (4.620)

Financial variable-0.004 0.028 0.006 0.009 0.051 0.001

(-0.300) (2.012) (1.249) (3.366) (7.941) (0.170)

Adj. R2 0.253 0.281 0.258 0.281 0.253 0.284 0.271 0.279

Tobin's q

qit 0.328 0.070 0.278 0.061 0.109 0.062 0.193 0.064(6.102) (6.162) (5.483) (5.465) (2.559) (5.617) (3.896) (5.520)

qit2 0.000 0.025 -0.007 0.020 0.028 0.019 0.014 0.022(-0.008) (2.365) (-0.210) (1.936) (1.053) (1.799) (0.450) (1.997)

Financial variable0.623 0.127 0.228 0.031 0.288 0.027

(16.723) (9.401) (38.772) (13.614) (22.310) (5.639)

Adj. R2 0.055 0.195 0.155 0.223 0.422 0.251 0.219 0.205

Profitability Shocks (residuals)

Mandated inv. Rate

Class 1: low dividend to capital ratio Class 2: high dividend capital ratioNone CF_K Sales_K WorkingK_K

Note: The estimation technique is OLS. 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), kit is the mandated investment rate, and qit is Tobin's q. 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. 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 dividend to capital ratio.

Table 8 Response of Investment to Alternative Measures of Fundamentals and Financial

Variables (When Firms are grouped by bond ratings)

Financial Variables

Class 1 Class 2 Class 1 Class 2 Class 1 Class 2 Class 1 Class 2(1) (2) (3) (4) (5) (6) (7) (8)

Profitability Shocks (first-order)

ait 0.249 0.127 0.233 0.093 0.203 0.090 0.220 0.122(20.767) (7.555) (19.764) (5.663) (16.115) (5.350) (18.393) (6.948)

ait2 0.202 0.268 0.180 0.239 0.188 0.272 0.193 0.267(4.782) (3.924) (4.374) (3.647) (4.515) (4.094) (4.668) (3.835)

Financial variable 0.065 0.461 0.026 0.050 0.050 0.021(6.548) (11.343) (10.545) (9.249) (13.148) (1.141)

Adj. R2 0.230 0.235 0.243 0.297 0.253 0.277 0.265 0.235

rait 0.131 0.108 0.127 0.038 0.078 0.057 0.101 0.107(14.420) (7.772) (13.273) (2.508) (7.901) (3.868) (11.057) (7.309)

rait2 0.109 0.182 0.092 0.150 0.076 0.152 0.078 0.193(5.503) (4.883) (4.768) (4.130) (3.897) (4.157) (4.028) (4.986)

Financial variable 0.032 0.456 0.033 0.050 0.055 0.023(2.728) (9.598) (12.206) (8.544) (13.675) (1.237)

Adj. R2 0.192 0.231 0.203 0.276 0.223 0.267 0.231 0.233

kit 0.198 0.181 0.198 0.140 0.185 0.151 0.170 0.187(22.937) (13.665) (21.984) (9.705) (16.726) (9.166) (18.275) (13.141)

kit2 0.067 0.079 0.066 0.073 0.065 0.078 0.061 0.087(3.936) (2.957) (3.952) (2.780) (3.777) (2.933) (3.593) (3.130)

Financial variable-0.009 0.308 0.006 0.020 0.036 -0.028

(-0.820) (6.774) (1.905) (3.002) (7.993) (-1.538)

Adj. R2 0.243 0.292 0.249 0.313 0.243 0.296 0.256 0.292

Tobin's q

qit 0.223 0.096 0.176 0.091 0.089 0.087 0.127 0.098(6.113) (5.805) (5.044) (5.803) (2.902) (5.456) (3.701) (5.697)

qit2 0.031 0.031 0.026 0.008 0.035 0.024 0.029 0.026(1.264) (1.570) (1.133) (0.403) (1.708) (1.256) (1.284) (1.335)

Financial variable0.478 0.425 0.176 0.047 0.213 0.084

(17.399) (12.248) (38.736) (10.378) (22.418) (5.531)

Adj. R2 0.051 0.262 0.128 0.333 0.339 0.315 0.171 0.282

Profitability Shocks (residuals)

Mandated inv. Rate

Class 1: firms without bond rating Class 2: firms with bond ratingNone CF_K Sales_K WorkingK_K

Note: The estimation technique is OLS. Time and firm dummies are included. t-statistics are given in the parenthesis. The dependent variableis the investment rate. ait is the profitability shocks (first-order), rait is the profitability shocks (residuals), kit is the mandated investment rate, and qit is Tobin's q. 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. While Class 1 firms are defined as the ones without debt rating, Class 2 firms have bond rating.

Table 9 Response of Investment to Alternative Measures of Fundamentals and Financial

Variables (When Firms are grouped by the debt to capital ratio)

Financial VariablesClass 1 Class 2 Class 1 Class 2 Class 1 Class 2 Class 1 Class 2

(1) (2) (3) (4) (5) (6) (7) (8)

Profitability Shocks (first-order)

ait 0.188 0.246 0.169 0.232 0.149 0.195 0.170 0.197(15.118) (16.242) (13.711) (15.630) (11.815) (12.135) (13.981) (12.542)

ait2 0.152 0.257 0.147 0.227 0.134 0.247 0.133 0.250(3.306) (4.757) (3.240) (4.316) (2.970) (4.649) (2.969) (4.691)

Financial variable 0.108 0.060 0.028 0.031 0.040 0.098(9.662) (4.191) (11.382) (8.585) (12.212) (11.023)

Adj. R2 0.280 0.192 0.305 0.201 0.314 0.214 0.319 0.228

rait 0.114 0.138 0.084 0.146 0.069 0.080 0.089 0.100(12.291) (11.636) (8.372) (11.830) (6.884) (6.135) (9.488) (8.251)

rait2 0.083 0.154 0.065 0.131 0.055 0.117 0.054 0.120(4.000) (5.693) (3.126) (4.974) (2.704) (4.370) (2.632) (4.463)

Financial variable 0.097 0.003 0.030 0.040 0.039 0.123(7.664) (0.161) (11.020) (10.080) (11.318) (12.976)

Adj. R2 0.256 0.162 0.273 0.173 0.290 0.194 0.291 0.214

kit 0.163 0.223 0.160 0.217 0.148 0.206 0.145 0.184(18.606) (19.537) (16.519) (18.865) (14.000) (13.514) (15.346) (14.908)

kit2 0.089 0.054 0.089 0.053 0.087 0.051 0.083 0.043(4.901) (2.460) (4.872) (2.495) (4.783) (2.309) (4.571) (1.944)

Financial variable0.009 0.003 0.007 0.008 0.020 0.078

(0.741) (0.214) (2.404) (1.712) (4.901) (8.734)

Adj. R2 0.303 0.225 0.303 0.231 0.304 0.226 0.309 0.247

Tobin's q

qit 0.087 0.320 0.076 0.276 0.075 0.097 0.074 0.162(6.738) (5.436) (5.985) (5.018) (5.946) (2.121) (5.781) (3.199)

qit2 0.016 0.078 0.013 0.062 0.014 0.090 0.014 0.083(1.833) (1.740) (1.539) (1.490) (1.585) (2.610) (1.545) (2.202)

Financial variable0.136 0.755 0.035 0.234 0.036 0.555

(10.155) (18.522) (12.923) (40.135) (9.377) (33.677)

Adj. R2 0.227 0.047 0.258 0.167 0.276 0.430 0.254 0.357

Profitability Shocks (residuals)

Mandated inv. Rate

Class 1: low debt ratio Class 2: high debt ratioNone CF_K Sales_K WorkingK_K

Note: The estimation technique is OLS. 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), kit is the mandated investment rate, and qit is Tobin's q. 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. Class 1 (Class 2) firms are defined as the lower (higher) 50 percentile of the firms when they are sorted by the size of the ratio of their total debt to the capital stock.

Table A1 - Appendix Number of Firms in 2-digit Industries

SIC Code Name Number of firms

20 Food and Kindred Products 3421 Tobacco Products 222 Textile Mill Products 1223 Apparel and other Textile Products 624 Lumber and Wood Products 1025 Furniture and fixtures 1426 Paper and Allied Products 1427 Printing and Publishing 2328 Chemicals and Allied Products 5229 Petroleum and coal products 930 Rubber and Miscellaneous Plastic Products 1331 Leather and Leather Proucts 332 Stone, Clay, and Glass Products 1133 Primary Metal Industry 1334 Fabricated Metal Products 2535 Industrial Machinery and Equipment 6636 Electronic and Other Electric Equipment 6737 Motor Vehiles and Equipment 3138 Instruments and Related Products 5239 Miscellaneous Manufacturing Industry 6

TOTAL 463

Table A2 - Appendix Calibration of the Parameters used in Calculation of the Mandated Investment Rate

Parameter Definition Description and data Source Average

value ψ elasticity

of capital with respect to its cost

2-digit industry level (panel regression of log of real profit to cost of capital for each industry) COMPUSTAT

-0.68

η 1/(1-α)

where α is cost share of capital

α = (rental price of capital * capital stock)/ total cost of production. Firm-level. Data source for rental price: BLS database Data source for total cost of production: COMPUSTAT

1.37

c Cost of

capital (rt+δt)(pit/pt)[Tt/(1-τ)] r: real interest rate δ: depreciation rate (2-digit form BLS database) (pit/pt): ratio of new capital deflator to shipment price index (4-digit Gary and Bartelsman data set) [Tt/(1-τ)]: tax parameter (2-digit BLS database)

0.16

Table A3 - Appendix Calibration of the Parameters used in Calculation of Profitability Shocks 1/

Parameter Definition Description and data Source Average

value αL Cost share

of labor Same for each firm. Ratio of wages times employment level to the sum of the rental price of capital times capital level and wages times employment level. Employment and capital data: COMPUSTAT. Rental price of capital: BLS database. Wage: Gray and Bartelsman 4-digit dataset.

0.73

θ Profit

function coefficient

Estimated coefficient by regressing profit function on capital Source: COMPUSTAT

0.61

ξ

Elasticity of demand

ξ = [(θ-1)(1-αL)]/ [αL (θ-1)+1] Same for each firm

-0.15

1/ These shocks are obtained from the first order condition for profit maximization with respect to labor.