Debt financing in private and public firms

Abstract

Using administrative confidential data on the universe of Canadian corporate firms, we compare debt financing choices of private and public firms. Private firms have higher leverage ratios, which are entirely driven by private firms’ stronger reliance on short-term debt. Further, private firms rely more of leverage during economic expansions, while public firms rely on equity financing. Specifically, private firms manage to increase their long-term debt during expansions, while short-term debt is used during downturns. Our findings have implications for a better understanding of the role of asymmetric information in private firms’ capital structure decisions.

Appendix

1.1 Data appendix

The data appendix provides details on the assets, equity, short-term debt, long-term debt, debt, profits, tangible assets and sales variables. The following information lists the components of each variable along with the corresponding tax line from the T2 tax forms in brackets. The information is taken from the general index of financial information (GIFI) on the T2 corporate tax return.

• 1. Total assets (2599) =

  • Total current assets (1599) [Cash and deposits (1000) + Accounts receivable (1060) + Allowance for doubtful accounts (1061) + Inventories (1120) + Short term investments (1180) + Loans and notes receivable (1240) + Other currents assets (1480)]

  • Tangible capital assets (2008) [Land (1600) +Depletable assets (1620) +Buildings (1680) +Machinery, Equipment, Furniture and Fixtures (1740) +Other tangible assets (1900)]

  • Intangible capital assets (2178) [Intangible assets (2010) (i.e., goodwill, quota, licenses, incorporation costs, trademarks/patents, customer lists, rights, and research and development). + Resource rights (2170)]

  • Long term assets (2589) [Due from shareholder(s) and/or director(s) (2180) + Due from members (2190) + Investment in joint venture(s) and/or partnership(s) (2200) + Long term investments (2300) + Long term loans (2360) + Other long term assets (2420)]

  • Assets held in trust (2590)

• 2. Total liabilities and shareholder equity (3640)

• 3. Total debt = Total liabilities (3499) =

  • Total short-term debt = Total current liabilities (3139)

  • Total long-term debt = Total long-term liabilities (3450)

• 4. Total shareholder equity (3620) =

  • Common shares (3500) +

  • Preferred shares (3520) +

  • Retained earnings/deficit (3600) [Retained earnings/deficit—Start (3660) + Net income/loss (3680) + Capital contributed at the beginning of the fiscal period (3690) + Dividends declared (3700) + Prior period adjustments (3720) + Other items affecting retained earnings (3740)]

• 5. Total sales = total revenues (8299)

  • Total sales of goods and services (8089) +

  • Investment revenue (8090) +

  • Interest income (8100) +

  • Commission revenue (8120) +

  • Rental revenue (8140) +

  • Vehicle leasing (8150)

  • Fishing revenue (8160) +

  • Realized gains/losses on disposal of assets (8210) +

  • NPO amounts received (8220) +

  • Other revenue (8230) (Income/loss on subsidiaries/affiliates + Income/loss on joint ventures + Income/loss on partnerships + Alberta royalty tax credits)

• 6. Profits = Net income for tax purposes (300)

All variables are measured in nominal terms within the database. We use the consumer price index data available through CANSIM (CANSIM Table No. 326-0001) to deflate values of these variables.

Accounting rules dictate that:

• total assets (2599) $\equiv$ total liabilities (3499) + total shareholder equity (3620) identity holds.

1.2 Definition: private versus public firms

We classify firms as private or public on the following criteria:

• 1. Canadian-controlled private corporation (CCPC):

  • Resident incorporated firm not directly or indirectly controlled by non-residents, a public corporation or any combination; or

  • a private, resident corporation not directly or indirectly controlled by one or more public corporations or Federal Crown corporation

• 2. Public corporation:

  • Resident in Canada and having a class of shares listed on a prescribed Canadian stock exchange; or

  • Any Canadian corporation controlled by a public corporation

1.3 Industry sales growth and volatility of unexpected firm sales growth

Following Castro et al. (2015), we use a two-step procedure to measure unexpected industry sales growth and its volatility. As the first step, a firm’s sales growth is decomposed into predicted and unexpected components using the following regression:

$$\begin{aligned} \ln ({ Sales}_{{ it}})= & {} \alpha _i + \beta _1 \ln ({ Sales}_{i,t-1}) + \beta _2 \ln ({ Sales}_{i,t-2}) + \phi _1 \ln { Firm\,age}_{{ it}}\nonumber \\&+ \,\tau d1984_{i} + \phi _2 [d1984_{i}\times \ln { Firm\,age}_{{ it}}] + \mu _{{ it}}, \end{aligned}$$

(A.1)

where the residual $\mu _{{ it}}$ captures the unexpected component to firm sales growth. The dummy variable d1984 takes the value one for firms that existed prior to 1984 because we do not know their exact age, and zero otherwise. In the second stage, both $\mu _{{ it}}$ and $\mu _{{ it}}^{2}$ are regressed against time-specific, two-digit industry dummy variables given in the following equation:

$$\begin{aligned} \mu _{{ it}}= & {} \sum _{i\in j}\sum _{t}\varPsi _{jt} + \varepsilon _{{ it}}, \end{aligned}$$

(A.2)

and

$$\begin{aligned} \mu _{{ it}}^{2}= & {} \sum _{i\in j}\sum _{t}\varGamma _{jt} + \nu _{{ it}}; , \end{aligned}$$

(A.3)

for firm i in industry j at time t. These regressions include a full set of industry-time dummy variables $\varPsi$ and $\varGamma$.¹⁵ $\varPsi _{jt}$ gives the unexpected firm sales growth within industry j at time t, while $\varGamma _jt$ gives the variance of the unexpected sales growth within industry j at time t. The estimated coefficients on the dummy variables in Eq. (A.2) capture the average unexpected sales growth within an industry at a given time (${\widehat{\varPsi }}_{jt}$), while the estimated coefficients on the dummy variables in Eq. (A.3) capture the volatility of the unexpected firm sales-growth at a given time and industry (${\widehat{\varGamma }}_{jt}$).