The purpose of this research is to examine the sensitivity of cash flow to investment when firms are financially constrained and unconstrained. Our findings are consistent with the theoretical prediction of investment cash flow sensitivity, which suggests that constrained firms have higher investment cash flow sensitivity than unconstrained firms. For example, a firm's internal cash flow can affect investment spending due to a "financing hierarchy," in which internal funds have a cost advantage over new debt or equity financing.
One issue that has received particular attention is the sensitivity of investments to cash flow. First, when the firm faces financial constraints, the firm has the constraint to access external funds, therefore, its investment will depend on the availability of internal fund which causes the high sensitivity of cash flow to investment. Based on the work of Fazarri et al. 1988), if firms are financially constrained, the sensitivity of investments to cash flow is affected by asset tangibility.
The multiplier suggests that the sensitivity of investment cash flow in the tangibility of companies should be increasing. We found that tangibility did not affect the cash flow sensitivity of financial unconstrained firms. We summarize that there is a greater sensitivity of cash flow to investment in constrained firms compared to firms that can freely access an external source of funds.
The implication is that the cash flow generated by the business should not influence an investment decision.
Empirical studies
Based on the perspective of an imperfect capital market when firms are financially unconstrained, which implies that firms do not have restrictions to access external funds the same as a perfect capital market, their investment decision is also based on investment opportunities , therefore internal cash flow should not influence an investment decision. In addition to Fazarri, Hubbard and Petersen (1988), they show that firms that have sufficient free cash flow may also decide to invest with investment opportunities. We can use growth as one of the factors to identify investment opportunities which is measured by the ratio of the market capitalization of the stock and debt to the total assets of the firms.
Kaplan and Zingales (1995) study a relationship between corporate investment and cash flow to test the presence and importance of financial constraints. They undertake an in-depth analysis of 49 low-dividend firms identified by Fazarri et al. 1988) as there is extremely high sensitivity of investment to cash flow. Based on their study, only 15% of firm-years have some question about the firm's ability to access internal or external funds to increase investment.
They disagree that firms with high financial constraints will have greater investment-cash flow sensitivity than firms with less financial constraints. They conclude that the higher sensitivity of investment cash flows cannot be interpreted as evidence that companies are more financially constrained. Hovakimian and Titman's (2003) paper examines the importance of financial constraints for fixed investment expenditures by looking at the relationship between investment expenditures and returns from voluntary asset sales in financially sound US manufacturing companies.
Specifically, they examine whether asset sales have a greater impact on investment spending for firms that are likely to be financially constrained. Asset sales may provide a cleaner indicator of liquidity than cash flow, as it does not appear to be positively related to future investment opportunities. They find that after controlling for investment opportunities and cash generated from operations, cash obtained from asset sales is an important determinant of corporate investment.
Furthermore, the sensitivity of investment to proceeds from asset sales is significantly stronger for firms that are likely to be associated with characteristics related to financial constraints. However, they find that tangibility has no effect on the cash flow sensitivity of financially unconstrained firms. This argument implies a non-monotonic effect of tangibility on cash flow sensitivity: at low levels of tangibility, investment sensitivity to cash flow increases with asset tangibility, but this effect disappears at high levels of tangibility.
METHODOLOGY
Data selection
Relationship model of investment, cash flow and asset tangibility
- Specification
- Model estimation
After sorting firms into restricted and unrestricted groups, Equation (1) can be estimated separately across those different categories. However, one of the central predictions of our theory is that the financially constrained status is endogenously related to the tangibility of the firm's assets. Therefore, we need an estimator that includes the effect of tangibility on both cash flow sensitivities and on the restricted status.
This model allows the probability of being financially constrained to depend on asset tangibility and variables used in the literature (eg firm size and growth opportunities). As explained next, the model simultaneously estimates the equations predicting the constraint status and the investment spending of constrained and unconstrained firms. The conversion of regression estimates allows for endogenous selection into ``financially constrained'' and ``financially unrestricted'' categories via maximum likelihood methods.
Investment regime 1 is classified as financially constrained, which investment may be more sensitive to the availability of internal funds than regime 2, which is classified as financially unrestricted. We compress the notation for brevity and let Xit = (Qi,t-1 , Cash Flow,t , Tangibilityi,t , (Cash Flow* . Tangibility)i,t) be the vector of exogenous variables and α is the vector of coefficients. Equation (4) is the selection equation that establishes the firm's probability of being in regime 1 or regime 2.
Where yit* is a latent variable that weights the probability that the firm is in the first or second regime. The extent to which investment expenditures differ across the two regimes and the likelihood that firms will be assigned to each regime are determined simultaneously. We note that to fully identify the switching regression model, we need to determine which regime is the restricted one and which regime is the unrestricted one.
The algorithm specified in Equations (2) - (5) creates two groups of firms that differ according to their investment behavior, but it does not automatically tell the econometrician which firms are constrained. One advantage of our approach is that it allows us to use multiple variables to predict whether firms are constrained or unconstrained in the selection equation (4). In particular, the estimation of the selection equation allows us to assess the statistical significance of a given factor assumed to be a proxy for financing constraints, while controlling for the information contained in other factors.
Variables
- Dependent variables: Investment
- Independent variables
The cash flow is the sum of the net profit and depreciation during the year, divided by the tangible fixed assets at the beginning of the period. Normally, if the borrowing capacity of the company is high enough, the company becomes financially unconstrained and the sensitivity to investment cash flow is low. This means that further changes in tangibility will not affect the investment-cash flow sensitivity of a financially unconstrained company.
Tangibility is defined as the book value of property, plant and equipment (PPE) divided by total assets, as in the research of Qiu and La (2010) and Campello and Giambona (2010). We use four different sets of instruments to study the relationship between cash flow and tangibility that affects capital expenditures. Log assets is the natural logarithm of total assets in thousands as a measure of firm size.
If the companies have more asset size, there is more opportunity to have financial unrestrict. The companies that have more long-term debt are more likely to have financial unrest. It is the ratio of the sum of the stock market capitalization and the book value of debt to the book value of total assets at the beginning of the period.
It equals the sum of cash and short-term investments divided by total assets at the beginning of the period. Companies with ample cash reserves do not have liquidity constraints as their investments are not limited by a lack of financing. Therefore, companies with more financial space are more likely to experience no restrictions.
RESULTS
For each of the estimation reports in Panel B, (one for each of the 8 proxies for Investment and Tangibility), the dependent variable is coded 1 for allocation in investment regime 1, and 0 for allocation in investment regime 2. Firms allocated in investment regime 1 are classified as financially restricted, and those allocated in investment regime 2 are classified as financially unrestricted. This classification is based on theoretical priorities about which firm's characteristics are likely to be associated with financial constraints.
Consider the results from the selection regression (Panel B), all variables classified firms in the unrestricted regime. The selection equation estimators show that firms that are larger and have higher long-term debt ratio, greater investment opportunities, higher level of financial slack are more likely to be financially unconstrained. Panel A explains the relationship of the independent variables (Growth, Cash Flow, Asset Tangibility and CF*Tangibility) with Investment.
Based on the results of the selection model in Panel B, the independent variables show little or no response to investment for unconstrained firms because the result of the statistical test is insignificant (P-value). In contrast, limited firms show significant coefficients in most estimates at the 1%, 5%, and 10% significant levels, respectively.
ROBUSTNESS CHECK
Regression model with ex ante constraint selection
CONCLUSION
APPENDICES
Appendix A Data
Appendix B Alternative selection variables
We then test in the specific samples that we think are the limited companies. As shown in Table B.2, we select five companies with constraints. Our selection criteria are a low long-term debt ratio and unknown companies to test. Fixed AVG_Payout Ratio AVG_tang1 AVG_tang2 AVG_tang3 AVG_tang4 AVG_Long_term.
