ARTICLE REPRINT JOURNAL OF EQUIPMENT LEASE FINANCING

Transcription

1 ARTICLE REPRINT JOURNAL OF EQUIPMENT LEASE FINANCING The Journal of Equipment Lease Financing is published by The Equipment Leasing and Finance Foundation. The Equipment Leasing and Finance Foundation is a non-profit organization affiliated with the Equipment Leasing Association of America. The Foundation s mission is to increase the body of knowledge in the equipment leasing and financing field N. Fairfax Drive, Suite 550, Arlington, VA 22203: Lisa A. Levine, Executive Director Phone: (703) Fax: (703) Website:

2 Leases, Debt and Taxes: A New Measure By James S. Schallheim, Ph.D. Anew way of measuring the tax rate, called the firm s marginal tax rate (MTR), suggests that firms with lower taxes are more likely to lease and that firms with higher tax rates are more likely to use debt financing. In addition to describing the MTR, this article discusses several factors that are characteristic of firms that favor leasing as well as factors that favor debt financing. What are the factors that favor leasing or debt financing? In a recent study published in the Journal of Finance, my colleagues and I demonstrate that certain factors favor leasing while other factors favor debt financing. 1 But the most striking evidence provided in our study is that a better and more precise way of measuring the tax status of the firm provides results that are consistent with finance theory and practice. One of the differences between the study and practice of lease financing is that practitioners often know from experience that certain relationships hold, while academics try to demonstrate these same relationships with large samples and statistics. One example of this dichotomy concerns the notion that low taxpaying firms favor lease over debt financing. While leasing companies advertise this advantage, academics have traditionally had trouble establishing this relationship between taxes and the propensity to lease or use debt. This article discusses a new way of measuring the tax rate, called the firm s marginal tax rate (MTR), based on the work of my colleague John Graham. Using this improved measure of the MTR, we are able to confirm the commonly understood relationship between lower taxes and the propensity to lease. Furthermore, we are able to show that firms with higher tax rates are more likely to use debt financing. Finally, we discuss several factors that are characteristic of firms that favor leasing, as well as factors that favor debt financing. Leasing companies should be able to use this information to promote leasing to firms displaying these leasing characteristics, or promote debt financing to firms favoring alternative characteristics.

3 One word of caution is in order. By no means are the factors derived in this study to be considered exhaustive. Rather, these are factors that are available from public information shown in the financial statements of public corporations. Industry practitioners are well aware that other factors influence the leasing decision by lessee firms. Some well-known examples exist, such as the use of capital and operating budgets by large corporations that sometimes makes it easier for a manager to acquire an asset through leasing rather than use a constrained capital budget for purchase. Another well-known example is the preference for off-balance sheet financing. For our data sample, it is not possible to separate such factors using the public accounting data. WHAT IS THE MARGINAL TAX RATE? The marginal tax rate (MTR) is simply defined as the tax rate that would apply to one additional dollar of taxable income earned by the firm (we focus on corporations). This tax rate can range from zero to the maximum corporate tax rate applicable at the time. Although easy to define, the actual computation of the MTR is not as simple. For example, the MTR is not determined by taking the income taxes or provision for income taxes from the income statement and dividing it by earnings before taxes. That computation might yield an average tax rate, but it would not determine the marginal tax rate. Consider the following example to illustrate the difference. 2 Suppose a firm earned $1 million last year but is expected to either earn $1 million or lose $3 million at the end of this year. The firm s combined federal, state, and local income tax burden is 40 percent on each dollar of profit and the firm pays no taxes on a loss. On the income statement for the year competed, the provision for income taxes is $400,000, reflecting a tax rate of 40 percent. But is 40 percent really the MTR for this firm? If we assume that the $1 million gain or the $3 million loss are equally likely by the end of the year, the firm is expected to lose $1 million. Expectation is calculated by the following formula: Expected outcome = Probability(1) x Outcome(1) + Probability(2) x Outcome(2) = 1 /2 x -$3 million + 1 /2 x $1 million = -$1 million This expected outcome could be interpreted as resulting in an expected tax rate of 0 percent in the future (one additional dollar of income would not be taxed). However, this argument also is incorrect. To determine the correct MTR for this example, we must compute the present value of the company s tax bill as it changes with an additional dollar of income today. Instead of earning $1 million in the profitable outcome, the firm would earn $1,000,001 and would pay tax of 40 cents on the extra dollar of income. If the firm loses $2,999,999 (as opposed to $3 million), the firm would pay no tax on the additional dollar of income. The expected tax bill on the additional dollar of current income is computed to be 20 cents (2 cents x 40 cents + 2 cents x 0 = 20 cents). The correct expected MTR is 20 percent for this firm, even though on average the firm was expected to suffer a loss at the end of the year. In fact, if we reverse the numbers in this example to an equally likely chance of the firm earning $3 million or losing $1 million, the MTR computation will be the same 20 percent. In that case, a firm that is expected to be profitable may still have a lower MTR (lower than 40 percent) if there is a reasonable chance that it could suffer a loss in the future. Determining the MTR is not simple. Furthermore, there are other factors that will be mentioned briefly that add even more complexity to the issue. Nevertheless, there is a very important implication here for leasing. That is, for the cases where the firm suffers losses, it would be better off leasing equipment because the firm would not benefit from some of its tax deductions such as depreciation. By allowing the lessor to take and fully utilize the depreciation deduction, as well as interest deductions from financial leverage, the lessee can benefit in the form of lower lease payments. The marginal tax rate is simply defined as the tax rate that would apply to one additional dollar of taxable income earned by the firm. J O U R N A L O F E Q U I P M E N T L E A S E F I N A N C I N G F A L L V O L. 1 8 / N O

4 Even large firms can be subject to the progressive portion of the tax schedule if their profits are small. FACTORS THAT DETERMINE THE MARGINAL TAX RATE 3 As illustrated in the examples above, the profitability of the firm is an important determinant of the MTR. Not just profitability, but the uncertainty associated with the level of income is critical to the determination of the MTR. Similarly, the uncertainty in the level of income can make a firm a more likely candidate for leasing for tax purposes.thus, as we will see, the lower the MTR, the more likely a firm will favor leasing. There are other factors affecting the MTR. One factor is the progressive nature of the tax schedule, which has greater impact on small firms (income less than $100,000). However, even large firms can be subject to the progressive portion of the tax schedule if their profits are small. Another important factor concerns firms that have net operating losses (NOLs). These NOLs can be carried back for two years and carried forward Figure 1 POSSIBLE REALIZATIONS OF TAXABLE INCOME ($ MILLIONS) Last Year Current Year Next Year Two Years in Future 100% Chance 100% Chance 50% Chance 50% Chance 50% Chance 50% Chance Path Path Path Path Figure 2 TAXES PAID ON AN ADDITIONAL DOLLAR OF CURRENT INCOME Last Year Current Year Next Year Two Years in Future Path 1 0 cents 0 cents 40 cents Tax Already Paid Path 2 0 cents 0 cents 40 cents Tax Already Paid Path 3 0 cents 0 cents 0 cents 40 cents Path 4 0 cents 0 cents 0 cents 0 cents for 20 years.this provision of the tax code can have a very large impact on the MTR calculation. For example, suppose that a firm has losses such that it uses up all its carryback options and a portion of its carryforward options. Suppose this firm will not pay taxes until five years from the current period (at which time it will pay 35 cents on a dollar of income). The present value of the firm s taxes owed, assuming a discount rate of 8 percent, would be computed as the present value of 35 cents in 5 years at 8 percent or 35 cents/(1+.08) 5 = 23.8 cents. The MTR for this firm would be 23.8 percent even though the firm has NOLs at the current time (which results in the carryback and carryforward). Other adjustments to the MTR computation include such items as state, local, and foreign taxes, and tax credits such as research and development, foreign, and investment tax credits. One major adjustment to the MTR applies to firms subject to the alternative minimum tax (AMT). The AMT often forces a firm to pay taxes when it otherwise would be in a zero taxpaying position. The alternative minimum tax rate is 20 percent, but there is also an AMT credit that could offset the AMT tax in future years. For leasing, the AMT has a special effect in that a firm could be subject to the AMT due to such tax preference items as accelerated depreciation, but the same firm might not be subject to AMT if it leased the depreciable equipment (as lease payments are not considered tax preference items). SIMULATING THE MARGINAL TAX RATE It should be clear that the computation of the MTR is very complex and can depend directly on the level of future unknown income levels. Therefore, how can the MTR be computed? John Graham derived a method of calculating the MTR based on simulation. 4 To understand how the simulation works, here is an example presented by Graham and Lemmon. 5 A firm has a current loss of $2 million. Last year it had income of $1 million. Next year and the 1 8 J O U R N A L O F E Q U I P M E N T L E A S E F I N A N C I N G F A L L V O L. 1 8 / N O. 2

5 following year it has a chance of earning $3 million or losing $1 million. The earnings profile appears in figure 1. Each path represents one possible way in which the earnings will evolve and has a 25 percent chance of occurring (.5 x.5 =.25). We will assume the firm s statutory tax rate is 40 percent on each dollar of profit. The firm has a loss of $2 million, which can be carried back for one year. It now has $1 million that can be carried forward, but the carryforward will depend on the possible future earnings along the four paths. For an additional dollar of taxable income, the tax paid on this additional dollar depends on the four paths and is represented in figure 2. For paths 1 and 2, the firm earns $3 million in the next year, which uses up the $1 million in tax loss carryforward and allows the extra one dollar of income to be fully taxed the next year. For path 4, the firm losses another $1 million next year and the year after that, allowing the extra dollar of income to remain untaxed for the entire period. For path 3, the firm has the $1 million in carryforward plus an additional $1 million loss in the next year for a total cumulative loss of $2 million to be carried forward to the year after that. Because the firm becomes profitable in the two-years-in-future outcome for path 3 and earns $3 million, the additional dollar of income becomes taxable at the full rate of 40 percent as reflected in the last column for path 3. This example illustrates two things. First, the MTR for this firm is 27.1 percent assuming an 8 percent discount rate (computed as:.5 x (.40/1.08) +.25 x (.40/1.082) =.271). Second, this type of problem can be solved using simulation techniques. By using a larger number of paths for up to 20 years in the future, a much more complete picture of the future tax status of the firm will result. In fact, such probability measures as the normal distribution (the bell shaped curve) have an infinite number of possible future paths. For the simulation technique used by Graham, 50 different future earning paths are computed (or drawn from a normal distribution), resulting in 50 different tax rates for each firm. The expected MTR for each firm is then the average of these 50 draws or simulations. Graham developed a computer algorithm that produces simulated tax rates for the 1980s and 1990s for over 10,000 firms. These estimated tax rates are available on the Internet. 6 These simulated tax rates are also used by Ibbotson Associates to estimate weighted average cost of capitals (WACCs) in their Cost of Capital Quarterly Yearbook. THE BEFORE-FINANCING MARGINAL TAX RATE The final factors affecting the MTR that will be discussed involve the financing decision and its impact on the MTR. Because we are looking for the impact of the MTR on the debt and leasing decisions of the firm, we must also consider the impact of the debt and leasing decisions on the MTR. This effect is simple yet extremely important. Suppose a firm has increased its use of debt in order to increase the interest deduction and thereby lower its tax bill. This firm will appear to have a lower MTR. Analyzing this firm for the impact of the MTR on the use of debt could easily lead to the wrong conclusion. Namely, it would appear that a firm that uses more debt, or increases its use of debt, is in a lower MTR position. In fact, the opposite is true: the higher MTR firms increase their use of debt. 7 How is it possible to control for the problem of the simultaneous effect of the financing decision on the MTR? We introduce a new measure called the before-financing marginal tax rate. The before-financing MTR is calculated by adjusting the taxable income for the financing effect that is, the interest expense is added back. For debt, this is a straightforward adjustment. For leasing, the adjustment is much more difficult. To approximate the interest component of the lease payment, we note the following. A lease payment is composed of several factors: interest (or return on capital invested in the leased equipment), depreciation (to recover the fall in the economic value of used equipment), and additional amounts to cover such items as trans- The before-financing MTR is calculated by adjusting the taxable income for the financing effect that is, the interest expense is added back. J O U R N A L O F E Q U I P M E N T L E A S E F I N A N C I N G F A L L V O L. 1 8 / N O

6 A firm increases its probability of facing financial distress as it increases its use of fixed-payment obligations that arise from debt and leases. action costs (cost of putting the lease together), contracting costs (both items that can be put into the contract as well as costs that cannot be easily written into a contract), and maintenance costs (if a service lease). Our rough approximation of the interest component of the lease payment is one-third. Admittedly, this is not correct, but it turns out that our main points are not sensitive to this approximation. We use the before-financing MTR to show that operating earnings of the firm are indeed an important factor for the firm s financing decision motivated by tax considerations. OTHER (NONTAX) CHARACTERISTICS OF LEASING AND DEBT POLICIES The focus of our research has been on the MTR. By no means would we want to dismiss the other factors that affect the firm s decisions to use leases or debt. One such critical factor concerns the possibility of financial distress. A firm increases its probability of facing financial distress as it increases its use of fixed-payment obligations that arise from debt and leases. Therefore, it is important that we attempt to control for this important factor. Furthermore, financial distress can have a different impact on debt financing as opposed to lease financing. The reason for this difference has to do with the way in which the bankruptcy laws treat leases as opposed to debt, even secured debt contracts. Within bankruptcy, if the lease is assumed by the debtor, and during the period pending assumption or rejection, the lessee is required to continue to make scheduled lease payments to the lessor, giving the lease priority on par with administrative expenses. 8 In contrast, bankruptcy proceedings grant the debtor a stay on the payment of most other financial claims, including those of secured debtholders, until the bankruptcy is resolved subject to requests for adequate protection by such secured creditors. Therefore, we can predict a differential effect between measures of expected financial distress costs and the use of debt or lease financing. Debt is expected to be lower with expected financial distress costs whereas leasing is expected to be positively related to expected financial distress. Another important nontax factor in the financing decisions of the firm is contracting costs. Contracting costs are types of transaction costs that relate to the use of different types of financial securities. For example, firms with lots of growth opportunities would find the use of fixed claims such as those associated with debt and leases to reduce their flexibility to pursue these growth opportunities. Or these growth firms may find the cost of financing with debt or leases to be prohibitively high. Firms that have a lot of fixed assets are often able to borrow more due to the collateral value of these assets relative to intangible assets. Naturally, the leasing contract is tied to a specific fixed asset and thus provides collateral. Thus we would expect to see firms with large growth options using less debt and leases, and firms with greater amounts of fixed assets to use more debt and leases. Leasing can be disadvantageous to firms subject to government regulation. The return for utility shareholders is calculated from the firm s capital base. Operating leases do not count as part of the capital base. Capital leases may or may not count as part of the capital base depending on the application of accounting rules by state regulators or rate commissions. Debt always counts as part of the capital base. Therefore, we expect regulated firms to use less leasing and more debt. The final nontax factor that we consider is the size of the firm. Size may represent many things. Larger firms can support larger amounts of debt than their smaller counterparts. This may be due to the fact that larger firms have more stable cash flows or that larger firms have better access to capital markets and it is easier and less costly for them to obtain external funding. Leasing, however, may offer an opportunity for smaller firms to obtain external funding. This could be true because leasing reduces the transaction and contracting costs (such as financial distress costs and providing collateral) for smaller firms with 2 0 J O U R N A L O F E Q U I P M E N T L E A S E F I N A N C I N G F A L L V O L. 1 8 / N O. 2

7 more uncertainty about their cash flows. Thus leasing should be inversely related to firm size. SUMMARY OF RESULTS The data for this analysis came from the COM- PUSTAT tapes, a product of Standard and Poor s Corp. The firms are almost all publicly traded companies in the United States, with the exclusion of most firms in the financial industry. The sample includes the years 1981 to 1992 and resulted in a sample size of over 18,000 firmand-year observations. (Many firms are repeated for several years in the sample, but some firms may be only represented once if insufficient data are available.) ating) lease financing, whereas high tax rate firms are more likely to use debt. 10 We also examined, with multiple regression, the other nontax factors that we could measure and that affect debt and leases. For these factors, capital and operating leases looked almost identical. For financial distress costs, we use an expected bankruptcy cost measure and the Altman Z- score. 11 For contracting costs, we use market-tobook ratio to capture firms with growth opportunities. 12 We also measure firms in a regulated industry as well as firms that hold more 0.45 Figure 3 AFTER-FINANCING MTR Debt is measured as long-term debt scaled by market value of the firm. Figures 3 and 4 show the after-financing MTR and before-financing MTR for the years 1982, 1992, and the entire period The before-financing MTR ranges between 0 and 46 percent in our sample with an average of 33.1percent compared to the after-financing MTR average of 27.0 percent. The exhibits also illustrate the clustering of firms at the high values and, to a lesser extent, at the low end of the distribution. Relative Frequency Debt is measured as long-term debt scaled by market value of the firm. Capital leases are also scaled by market value. 9 Operating leases are computed as the present value of the next five years of rental commitments (discounted at 10 percent) and scaled by the firm s market value. We argue that operating leases are more likely to be true or tax-transfer leases whereas capital leases are likely to be more like debt, although they could be a mixture of true leases and conditional sales contracts. For our important variable of the simulated before-financing MTR, our results demonstrate that the relative levels of debt increase with the MTR, the relative levels of capital leases are not reliably correlated with the before-financing MTR, while the operating leases are negatively related to the before-financing MTR. These results demonstrate quite significantly that lower tax rate firms are more likely to use (oper- Relative Frequency % 5-10% 10-15% 15-20% 20-25% 25-30% 30-35% 35-40% 40-45% >45% Percent Figure 4 BEFORE-FINANCING MTR 0-5% 5-10% 10-15% 15-20% 20-25% 25-30% 30-35% 35-40% 40-45% >45% Percent J O U R N A L O F E Q U I P M E N T L E A S E F I N A N C I N G F A L L V O L. 1 8 / N O

8 Firms with higher bankruptcy costs favor the use of leases and discourage the use of debt. collateral. Finally, size is a measure of the firms market values. 13 In figure 5, the impact of these factors is summarized by an up or down arrow indicating that debt or leases increase or decrease with the given factor. Thus we find that firms with higher bankruptcy costs favor the use of leases and discourage the use of debt. Firms with more growth opportunities, as captured by the market-to-book ratio, are using less fixed-cost financing. Firms subject to regulation, such as utilities, favor debt financing over leases. Collateral helps support both debt and lease financing. Finally, larger firms appear to favor debt financing while smaller firms favor lease financing. firm that is more likely to use debt financing has a higher before-financing MTR (the beforefinancing qualification is very important), a lower expected cost of financial distress and bankruptcy, and a larger size, and it might be in a regulated industry. Firms with more growth opportunities avoid both debt and leases, whereas firms with higher collateral values are more likely to use both debt and leases. The author s biography may be found on page 36. Endnotes 1 Graham, John, Michael Lemmon, and James Schallheim, Debt, Leases, Taxes and the Endogeneity of Corporate Tax Status, Journal of Finance (February 1998): CONCLUSIONS How can this study help the leasing industry? First and foremost, the marginal tax rate, or MTR, is a very important factor in determining a lessee s choice of debt or lease financing. A new way to calculate the MTR was discussed in this article. These MTRs are available over the Internet. This article also provides a profile, based on publicly available information, for a firm that is more likely to lease or more likely to use debt financing. A firm that is more likely to lease has a lower MTR, higher expected costs of financial distress and bankruptcy, and a smaller size. A Variable Debt Leases Financial Distress Expected BK Costs Altman Z-Score Contracting Costs Size Market-to-Book Ratio Regulated Industries Collateral Figure 5 IMPACT OF OTHER NONTAX FACTORS 2 This discussion is based on an example presented in Measuring Corporate Tax Rates and Tax Incentives: A New Approach, by John Graham and Michael Lemmon, Journal of Applied Corporate Finance (Spring 1998): Ibid. 4 Graham, John, Debt and the Marginal Tax Rate, Journal of Financial Economics (May 1996): 41-73; and Graham, John, Proxies for the Corporate Marginal Tax Rate, Journal of Financial Economics (October 1996): Graham and Lemmon, Measuring Corporate Tax Rates. 6 or 7 The problem just described is referred to as the endogeneity effect of the MTR. In other words, the choices of debt and leasing is an endogenous decision made in conjunction with the other decisions of the firm, including its tax position. 8 In 1994, Bankruptcy Code section 365(d)(10) was added, requiring the debtor (lessee) to timely perform its lease obligations arising 60 days after filing, including payment of the contract rent pending assumption or rejection. An article that provides more detailed evidence on leasing and financial distress is Bankruptcy Costs and the Financial Leasing Decision, by V.S. Krishnan and Charles Moyer, Financial Management, Vol. 23, No. 2 (Summer 1994), Certainly there are problems with our measurement of the operating leases. For example, rental commitments also include capital leases, thus overstating the operating leases. But operating lease payments beyond five years are not captured by the COMPUS- 2 2 J O U R N A L O F E Q U I P M E N T L E A S E F I N A N C I N G F A L L V O L. 1 8 / N O. 2

9 TAT database, thus understating the operating leases. Corrections to these errors were recently published in L. Shanker s A New Approach to the Effect of Taxes on the Leasing Decision, Financial Practice and Education (Fall/Winter 1999). Nevertheless, the main conclusions of our article remained unchanged after correcting for the measurement error of operating leases. 10 As a way to check our result, we compare debt, capital leases, and operating leases to the after-financing MTR. Now both capital and operating leases are quite negatively related to the MTR, but so is debt. This result demonstrates that it is extremely important to correct for the endogeneity problem. 12 The market-to-book ratio is the market price per share of stock divided by the book value per share of common stock. It is often the case that firms with lots of growth opportunities have higher market-to-book ratios. 13 The measurement of these variables is very detailed and would require too much space for this article. Therefore, the interested reader is referred to the Journal of Finance article cited in the first endnote. 11 The Altman Z-score is a statistical measure of bankruptcy probability. The lower the Z-score, the higher the probability is that a firm will go bankrupt in the next year. Guidelines for Contributors Journal articles are encouraged from leasing professionals, the academic community, and other volunteers whose businesses relate to equipment leasing. Articles are intended to provide analysis of: legal, accounting, and regulatory developments; market segments; finance sourcing and trends; and current research in the field. A peer review process is employed for each article submission. A blind review system, in which the author s identity is not known to the editorial review board, is employed if the author so requests. Manuscript Format and Style Articles may range from 6 to 20 double-spaced pages, including figure and chart material. Number each page and figure or chart. Please submit 2 copies. All manuscripts will be edited to the Journal s style. Articles should not promote or reference the author s company or business but should discuss issues and topics as they relate to particular segments of the leasing industry. Computer Disk In addition, please include a 3.5-inch computer disk in Word Perfect or Word. Tabular material should also be in a Word Perfect, Word or ASCII format. Tables and figures should be prepared and placed on disk in a TIFF or EPS file format, if possible, along with a supplied hard copy version. Illustrations Reference all charts, tables, and figures in text (e.g., See figure 3), and provide adequate interpretation in text. Number these illustrations consecutively and give them titles. Denote their source at the bottom of each illustration. Authors are responsible for the accuracy of all illustrations. Endnotes Please group all footnotes at the end of the article, as endnotes. If appropriate, you may add or substitute a bibliography. References should identify all authors of a work, and include, at minimum, (1) titles (of books, chapters in books, articles, or special works), (2) city of publication, (3) publisher, and (4) year of publication. For Further Information or to mail queries and submissions, contact the managing editor as listed on the inside front cover of this journal. J O U R N A L O F E Q U I P M E N T L E A S E F I N A N C I N G F A L L V O L. 1 8 / N O