Do Measures of Financial Constraints Accurately Identify Constrained Firms?
Do Measures of Financial Constraints Measure Financial Constraints?
Farre-Mensa, Ljungqvist 2014
How financial constraints affect firm behavior is central in finance study, but we have relied on indirect proxies or on some popular indices. This paper finds these measures do not well identify firms that are constrained.
Two definitions prevalent in the literature:
1) the more inelastic the supply of capital, the more costly the firm finds it to raise additional external capital (on Stiglizt and Weiss, and Whited and Wu). 2) The larger the wedge, the more constrained a firm is (on Fazari, Hubbard, and Petersen)
On debt:
Using natural experiment, if corporate tax rate increases the firm finds tax benefit. If a firm faces a highly inelastic supply of debt, it should be unable to increase its leverage in response to a tax increase. (definition 1)
If a firm is constrained under the wedge definition (def 2), the greater the wedge between its internal and external costs of debt the more value the firm transfers to debtholders when issuing debt. So, it should raise less debt in response to a tax increase.
On equity:
If a firm faces constraints in the equity market (inelastic supply of equity), it should never engage in equity recycling. With the 2nd definition the firm facing the inelastic supply of equity the tendency to recycle equity should decrease in the size of the wedge.
Test results on the widely used measures of financial constraints say constrained firms do not in fact face an inelastic capital supply curve, nor do they appear to face a systematically larger wedge between their internal and external capital. In out of sample test in which the authors use Merton’s distance to default measure, both private and public firms are unresponsive to tax increases and do not engage in equity recycling.
Those measures identify young and fast growing firms rather than constrained firms. The findings suggest that ex-ante studies using those measures may reflect differences in the growth and financing policies of firms at different stages of their life cycles.
Measures of Financial Constraints:
Existing proxies try to infer financial constraints from firms’ statements about their funding situation or changes in investment plans, their actions, or their characteristics. The most popular one is the KZ index that challenges FHP 1988 in which investment is sensitive to cash flow and the sensitivity is the greatest among non or dividend paying firms. KZ show those firms are actually the least constrained firms. Lamont, Polk, and Saa-Requejo 2001 established the KZ index by using the coefficients on MB(+), leverage(+), CF(-), dividends (-), and cash(-). A higher index implies more constrained firms. It assumes prevalence of financial constraints varies neither over time nor over business cycle. Hadlock and Pierce 2010 update the KZ index by using size(-), size-squared(+), and age(-).
Next approach is to treat firms without a credit rating as constrained. Those firms cannot access the bond market and have to borrow from banks. And the rating itself reduces information asymmetry (Faulkender and Petersen 2006). Whited and Wu (2006) take another approach and composite index by using coefficients of CF to assets (-), a dummy of paying or non-paying (-), long term debt to total assets (+), size(-), sales growth (-), and industry sales(+). They get these coefficients by projecting shadow price of equity capital on to these variables.
Sample and Data:
Sample firms are all US firms traded in public markets. Table 1 shows in the above 5 rows the KZ index correlates the least with the other measures, and the HP and WW indices agree the most. In the last 5 rows, except for the HP and WW, there is somewhat lower agreement between the measures.
Summary stat: except the KZ index measures classify similar kinds of firms as constrained: younger, smaller, carry more cash on their balance sheets, have few tangible assets, lower return on assets, and lower marginal tax rates, less levered, rely more on short-term debt, and more likely to face a zero marginal tax rate. Those firms also have higher MB ratios and faster growth in sales and employment. They tend to invest significantly more in R&D. Bottom line is that those seemingly constrained firms do not appear to impede fast growth, R&D, or investment. Now, on the KZ index, the constrained firms do not show the above characteristics; instead they hold less cash, more tangible assets, and have higher leverage. And their MB ratios, growth in sales or employment are lower, while they invest more in fixed assets. They spend less on R&D.
Do measures of Financial Constraints Measure Financial Constraints?:
The authors want to test whether the commonalities shown by the measures except the KZ are driven by financial constraints or they reflect some other differences (life cycle stage). They focus on constrained firms’ ability to raise debt and equity.
Financial constraints and the curvature of the capital supply curve:
Simply, the firm would not receive a loan even if it offered to pay a higher interest rate. This happens when the supply curve is too stiff, or even vertical.
Financial constraints and the wedge between the internal and external costs of funds:
FHP 1998 argue when there is a positive wedge between internal and external capital, the firm is financially constrained. However, KZ 1997 contend even a small transaction cost of raising external capital would put a firm into the constrained category. Thus, this notion is only useful as a means to rank firms. In this case, a firm can be defined financially constrained though the supply curve is elastic.
Discussion:
Two graphs which show the constrained and non constrained cases seem important. Since it is challenging to estimate a firm’s opportunity cost of internal funds, the literature attempts to measure the constraints indirectly (what manager says). It is only valid if managers’ words and actions reflect the shape and location of the curve. The authors test take these endogenous choices as predetermined and try to assess whether, conditional on the choices, the behavior of constrained firms is consistent with being constrained.
Debt test:
The trade-off theory says a firm’s demand for debt should increase in its marginal tax rate. The authors take state corporate income tax rates as plausibly exogenous shocks to a firm’s demand for debt.
Debt test:
On the curvature definition – null hypothesis is constrained firms cannot increase leverage after the shock. On the wedge definition – only the magnitude of increase differs.
Identifying assumptions:
Credit market could matter if states increase tax rates in a boom, or if the supply of capital increased procyclically. Heider and Ljunqvist find no evidence for that. And the authors use diif-in-diff to mitigate the issue.
Empirical specification:
Restrict the sample to firms headquartered in states with a tax increase and their immediate neighbors headquartered in adjacent states without tax changes. This minimizes the impact of unobserved differences in local economic conditions.
Results on debt test:
Firms classified as constrained have no difficulty increasing their leverage in response to tax increases. The estimated supply curves, coefficients on the measures, are remarkably flat. On the wedge, none of the measures identifies as unconstrained firms that increase their leverage by more in response to a given tax increase than constrained firms.
Equity test: equity recycling
According to the literature, equity recycling occurs since investors force managers to pay high payouts so that firms are often forced to raise external capital to finance investments. By this scrutiny, investors minimize the risk that managers invest suboptimally. Or, firms could time the market (over and under valuation). Key is if a firm can raise and payout equity capital, it does not face an inelastic supply of equity so it is not constrained. On the wedge definition, just like the debt, constrained firms should payout a smaller fraction of their equity issuance proceeds.
Identification:
On the wedge definition, a firm’s propensity to engage in equity recycling to be decreasing in the wedge. On the curvature definition, firms with an inelastic equity supply curve, there would be no equity recycling at all.
Empirical specification:
Just like the debt test, first diff OLS regression.
Results on equity test:
Supposedly constrained firms do engage in equity recycling. And constrained firms recycle no less than unconstrained firms for three of five measures. When dividends and KZ measures are used, constrained firms do recycle significantly less. This seems possible because these two measures load heavily on dividends. But an additional test shows the dividends do not simply drive the test result.
Privately held firms and firms close to default:
Logic is firstly privately held firms are more likely to be financially constrained than public firms. The authors find private firms face financial constraints in debt and equity market, and the effect is the strongest for smallest private firms. To identify truly constrained firms the authors use Merton’s distance to default measure. They find like small private firms, but unlike public firms classified as constrained by traditional measures, firms that are close to default behave as if they faced inelastic debt and equity supply curves and thus were constrained according to both curvature and wedge definitions.
Difference between supposedly constrained firms and unconstrained ones
Constrained firms are more likely to fund themselves by issuing common equity. And they (according to KZ, HP, and WW) similarly raise equity more frequently than unconstrained firms. They rarely issue bonds but, regularly access the syndicated loan market (private). This indicates that the conventional measures of financial constraints do not generate a random partition of public firms. Plausible interpretation would be these measures identify firms in the growth phase of the life cycle.