Trade Receivables Policy of Distressed Firms and Its Effect on the Costs of Financial Distress

Beginning of article

This paper studies the trade receivables policy of distressed firms as the trade-off between the firm's willingness to gain sales and the firm's need for cash. We find that firms increase trade receivables when they have profitability problems, but reduce trade receivables when they have cash flow problems. We also find that a firm that significantly cuts its trade receivables when in financial distress will experience an additional drop of at least 13% in sales and stock returns over the previously documented 20% average drop for financially troubled firms. Moreover, the performance decline of a firm in financial distress is significantly higher if the firm cuts trade receivables than if it does not.

**********

Trade receivables are a large part of firms' assets. Mian and Smith (1992) report that 21% of the total assets of US manufacturing firms in 1986 were invested in financing clients. Deloof (2003) documents that 17% of the total assets of Belgian firms in 1997 were account receivables. The management of trade receivables, given its importance for firms' assets, has the potential to play an important role when firms encounter financial problems. Previous studies have focused on the estimation of the costs of financial distress (Altman, 1984; Alderson and Betker, 1995; Andrade and Kaplan, 1998; Molina, 2005), in some cases explicitly recognizing the importance of the relations with clients for capital structure decisions and for the costs of financial distress (Titman, 1984; Opler and Titman, 1994). One question that has not been considered before, however, is how troubled firms, and the costs of their financial distress, may be affected by their trade receivables policy.

In this paper, we address two questions. First, we study the trade receivables policy of a firm in financial distress as the trade-off between the firm's willingness to gain sales by financing its clients' purchases and the firm's need for cash. Second, we measure the effect of suboptimal trade receivables investment policies on the costs of financial distress. Financing clients via trade receivables can be seen as a short-term investment to capture clients, and we know that firms in financial distress are expected to underinvest. (1) Consistent with this intuition, Mian and Smith (1992) find that firms with lower bond ratings increase the use of factoring to manage their accounts receivables, suggesting that they are willing to collect their receivables faster as the quality of their ratings decreases. In contrast, Petersen and Rajan (1997) find that firms whose sales drop and firms with negative profits increase trade receivables to their clients. They argue that this increase might be due to a voluntary attempt to gain market share and sales or to an unwanted increase in receivables given the impaired ability of troubled firms to enforce the timely collection of their commercial credit. If this last interpretation is correct, the increase in receivables could be considered a cost of financial distress. Trying to buy market share by extending additional financing to clients may seem appealing to a troubled firm, as Petersen and Rajan (1997) suggest, but this strategy can be very costly, especially for those firms whose access to financial credit is severely curtailed.

To reconcile these two seemingly contradictory views, we explore the nature of the financial distress problem in greater detail by defining it in two different stages: 1) firms facing profitability problems, usually at the prefinancial distress stage and 2) firms facing cash flow problems, usually in full-blown financial distress. We study and compare the trade receivables policy of firms in both groups. We argue that firms facing profitability problems may attempt to apply an aggressive credit policy to clients in order to gain market share, especially if they have the market power to do so without incurring significant sales losses. Firms facing cash flow problems, however, should try to decrease their investment in clients' credit in order to get cash, especially if they can afford to do so without relinquishing an excess of their sales volume to their competitors.

Our results suggest that firms: 1) tend to increase the use of trade receivables when they start facing profitability problems, usually in a prefinancial distress situation and 2) provide fewer trade receivables to their clients when they face cash flow problems and enter full financial distress. Our results support the hypothesis that firms might try to buy market share when they face profitability problems but cut their trade receivables in an attempt to get cash when they experience serious cash flow problems.

However, it would seem that only firms that can exert market power are likely to succeed in buying market share by increasing trade receivables and obtaining cash by reducing the terms of trade receivables without paying a large penalty in terms of a sales drop. Therefore, firms in competitive industries may find it difficult to pursue either of the two strategies in a cost-effective manner. (2) Supporting this hypothesis, we find that firms in concentrated industries, which are assumed to have higher market power, tend to show a larger effect of financial distress on trade receivables; they show larger increases in trade receivables when facing profitability problems, and larger reductions in trade receivables when facing cash flow problems in financial distress.

We also study the effect that a decrease in trade receivables has on the performance of firms in financial distress. Our results are consistent with the drop in performance for firms in financial distress documented in the literature, but we add to this body of knowledge by demonstrating that the drop is significantly larger when there is a reduction in trade receivables. A firm that experiences financial distress will have a drop in sales of about 20% to 28%, but if the firm decreases its trade receivables by an amount larger than the 10th percentile of the sample, sales will drop an extra 13% to 20%. In other words, decreases in trade receivables account for at least one-third of the drop in performance of firms in financial distress.

To complement the previous analysis, and overcome potential fears about a structural endogeneity, we use a setting very similar to the one used by Opler and Titman (1994). We also examine the additional costs of financial distress for firms with high leverage that, following an industry downturn, decrease their trade receivables. Our findings support the idea that trade receivables management is important for financially troubled firms. Highly leveraged firms in situations of economic distress experience a significantly higher drop in sales if they cut their trade receivables.

This paper contributes to the financial distress literature in at least two ways. First, we assess the trade receivables policy of troubled firms, which helps to explain the role that trade receivables play when firms are in financial trouble. Second, we explicitly provide an estimate for the cost of decreasing the investment in trade receivables when firms face financial distress.

The paper proceeds as follows. Section I describes the data sample. Section II explains the empirical strategy and studies the trade receivables policy of firms in financial distress. Section III discusses the importance of the industry structure. Section IV analyzes the trade receivables policy of firms in predistress circumstances. Section V revises the effect of cutting trade receivables on the costs of financial distress. Section VI presents the concluding remarks.

I. Data

The sample considers firms in Compustat for which trade receivables data are available for the 1978-2000 period. We drop all firms with net sales lower than $5 million, firms that do not report positive cost of goods sold, and firms with total assets lower than $10 million. We also discard all companies in the banking, insurance, real estate, and trading industries (Standard Industrial Classification [SIC] codes between 6000 and 6999), and the nonclassifiable establishments (SIC between 9995 and 9999). Additionally, we drop all the firms in the services industries according to the Fama and French (1997) classification (SIC between 7000 and 8999). (3) The total number of firm-year observations for which our dependent variable is not missing from 1978 to 2000 is 79,926; obviously, missing observations in other variables diminish the number of observations in our regressions. We have been conservative in our approach to removing outliers in order not to affect the evidence of firms in financial distress; we eliminated only the most extreme observations in each variable. (4) Table I presents the descriptive statistics of our data set, reporting the mean, standard deviation, and the 25th, 50th, and 75th percentiles of the main variables used throughout the paper.

Figure 1 plots the yearly mean of the ratio of trade receivables to sales (measured in days) for all the firms in the database from 1978 to 2000. The average number of days that firms finance their clients via trade receivables has grown overall during the 23-year period, but with a high dispersion around the mean.

II. The Effect of Financial Distress on Trade Credit

In this section, we analyze the behavior of firms' trade receivables when they enter financial distress, and consider the hypothesis that troubled firms decrease their investment in trade receivables in an attempt to get cash. We estimate the following equation to study the effect of financial distress on trade receivables:

[(TR/Sales).sub.it] = [[alpha].sub.i] + [beta][FD.sub.it-1] + [gamma][X.sub.it] + [[epsilon].sub.it]. (1)

In this model, TR/Sales is the ratio of trade receivables to sales, measured in days, and defined as TR/Sales = (Trade Receivables/Net Sales) x 360. (5) Here, FD is a dummy variable equal to one if the firm is in financial distress in a particular year, and zero otherwise, and X is a matrix of controls.

[FIGURE 1 OMITTED]

Given the lack of a widely accepted definition of financial distress, we consider three different measures that have been used in the literature. Our first approach, which follows Asquith, Gertner, and Scharfstein (1994), considers a firm to be in financial distress if its coverage ratio (defined as earnings before interest, taxes, depreciation, and amortization [EBITDA]/Interest Expenses) is less than one for two consecutive years or if it is less than 0.8 in any given year. Firms that are classified as being in financial distress are identified with a dummy variable named FINDIST.

Although FINDIST is one of the most common definitions for financial distress, it can capture firms that fail to meet the specified coverage ratio because: 1) the interest payments are too high and 2) the EBITDA is too low due to poor economic performance, even if the firm is not excessively leveraged. To account for this fact, we also use a second, and seemingly stricter, measure of financial distress that takes into account the leverage of the firm relative to its industry. We use a dummy variable, FDLEV, that is equal to one if the firm is both highly leveraged and financially distressed according to our first definition (i.e., FINDIST = 1), and zero otherwise. A firm is considered to be highly leveraged when its leverage is in the top two deciles of its industry in a particular year. (6)

We follow DeAngelo and DeAngelo (1990) to build our third definition of financial distress. We use LOSSFD, a dummy variable that is equal to one if the firm has three consecutive years of fosses, and zero otherwise. More specifically, a firm is considered to be in financial distress in the third year of losses if net profit, the first lag of net profit, and the second lag of net profit are negative. This measure is based on the intuition that a firm with three consecutive years of losses is likely to behave as a financially distressed firm, even in the absence of high leverage.

Our model includes a matrix of control variables: the firm's level of trade credit received from suppliers (i.e., TrPay/CGS), the firm's leverage, the lagged firm's sales growth (i.e., [DELTA][Sales.sub.t-1]), and the level of inventories scaled by costs of goods sold (i.e., Inventory/CGS). (7)

We estimate Equation (1) using a fixed effects model. (8) The results are in Table II. Column 1 shows the results using FINDIST as the measure of financial distress (our base case), while Columns 2 and 3 present the results obtained using FDLEV and LOSSFD as measures of financial distress. The coefficients for the financial distress variables are negative and significant in all models, suggesting that firms in financial distress reduce the level of investment in financing their clients via trade receivables. This result is consistent with the hypothesis …