# Pass-Through and C Corp Outputs under TCJA

## Abstract

**:**

## 1. Introduction

_{L}). The CSM formulations for G

_{L}include borrowing costs that increase with debt and allow for identification of the maximum G

_{L}(max G

_{L}) and thus the maximum firm value (max V

_{L}), since max V

_{L}equals max G

_{L}plus unlevered equity value.

_{L}), which is a function of interest payments and earnings retained for growth. This innovation allows us to test different historical and future projected annual growth rates for the same target credit rating. In this paper, we test three annual growth rates. First, we test a historical sustainable growth rate of g

_{L}= 3.12%, which is consistent with the annual compounded growth in real US GDP for the 70 years prior to TCJA with GDP data supplied by the US Bureau of Economic Analysis (2020). This usage assumes GDP is a result of the growth in businesses and, in particular, in the risk-taking residual equity ownership of businesses. Second, we test g

_{L}= 3.90%, which is consistent with the Tax Policy Center, TPC, (2018) that cites sources predicting that TCJA, on average, will increase GDP by about 0.8%. Thus, and as also pointed out by Hull (2019), a g

_{L}of 3.90% is consistent with a growth rate of 3.12% increasing by about 0.8%. Third, we test g

_{L}= 4.50%, which represents the two growth projections given by TPC that are at the high end. In regard to these two high end projections, TPC reports that the Congressional Budget Office (CBO) estimates the effect of TCJA will be an increase of 0.7% per year in GDP for the next ten years, while the Tax Foundation Taxes and Growth model projects an increase of 2.1%. The average of these two rates is 1.4%. This suggests a rate of about 4.50% when 1.4% is added to the long-run growth in real US GDP of 3.12% that occurred prior to TCJA. While these are projections for only ten years due to possible termination of certain TCJA provisions (especially for PTs), our tests assume permanency in these provisions. Unlike other key provisions, the drop to a flat corporate tax rate of 0.21 caused by TCJA (where the former maximum tax rate for CCs had been 0.35) is not set to expire.

ODVs for PTs are smaller than CCs and typically come with a slightly higher quality rating.

_{L}outputs when growth is 3.12%. For each $1,000,000 in before-tax cash flows, we find max V

_{L}for a PT is $10.390 M (M = million) for the pre-TCJA test and $10.758M for the TCJA test. For CCs, max V

_{L}is $9.476 M for the pre-TCJA test and $11.413M for the TCJA test. Whereas the increase in max V

_{L}for PTs is only 3.54% due to TCJA, the increase for CCs is 20.44%, which reflects the more favorable tax treatment for CCs under TCJA. When we compare pre-TCJA nongrowth max V

_{L}and growth max V

_{L}using a growth rate of 3.12%, we find that the PT growth max V

_{L}is slightly greater. In contrast, the pre-TCJA nongrowth max V

_{L}for CCs is comfortably larger than its pre-TCJA growth max V

_{L}. The latter outcome occurs because high CC tax rates in the pre-TCJA environment made internal growth unaffordable for a typical CC as internal funds (retained earnings) faced a high pre-TCJA corporate tax rate. When we repeat this comparison using the lower tax rates under TCJA for CCs, we now find the growth max V

_{L}is greater than the nongrowth max V

_{L}. When we test greater growth rates of 3.90% or 4.50% that are projected to occur under TCJA, we find the growth max V

_{L}outputs for both PTs and CCs are now much greater than those for nongrowth. We show that the average pre-TCJA max V

_{L}from all PT tests is $10.380 M, which is greater than the average of $9.754 M found for CCs. However, under TCJA, we find this advantage for PTs is reversed, as the average max V

_{L}of $11.802 M for CCs is greater than that of $11.163 M for PTs. This advantage slightly rises if we repeat our tests and use spreads for either 2017 or 2018 but lessens if we test lower effective tax rates. We conclude:

CC firm valuation increases by much more than PTs under TCJA with one reason being that the relative lower tax rates under TCJA for CCs makes growth more affordable for CCs compared to PTs. This increase makes switching from a PT to a CC profitable under TCJA. If growth increases as projected under TCJA, PTs and CCs will both profit substantially.

_{L}) values compared to CCs. Second, PTs also have lower values for the maximum percentage increase in unlevered equity (max %∆E

_{U}). Third, whereas PTs have lower values for the net benefit from leverage (NB) for the two pre-TCJA and nongrowth TCJA tests, we find that PT have greater NB values for the three TCJA growth tests, indicating that PTs add more value per dollar of new debt for TCJA growth tests. The values for leverage gain outputs can change as annual spread data changes. Robust tests using spreads for 2018 and lower tax rates generate lower values. The lower values for max %∆E

_{U}are consistent with empirical research. We conclude:

While the superiority of PT compared to CCs depends on the leverage gain output being analyzed, CCs generally perform much better in the category of leverage gain outputs indicating greater gains from leverage compared to PTs.

_{L}and V

_{L}results for PTs when plotted against respective DVs and credit ratings. Section 6 gives results pertaining to optimal outputs for PTs and CCs, provides policy implications, and suggests possibilities for future research. Section 7 offers a summary.

## 2. Literature Review

#### 2.1. Ownership Forms

#### 2.2. Financing Forms

#### 2.3. Effective Tax Rates

#### 2.3.1. Effective Tax Rate Described

#### 2.3.2. Effective Tax Rates for Main Tests

_{U}) retired with debt (D). Hull argues the corporate tax rate (T

_{C}) and the personal equity tax rate (T

_{E}) decrease with greater debt-for-equity transactions while personal debt tax rates increase. While Hull’s original arguments were for a CC, they are applicable to PTs since T

_{E}for PTs moves in the same direction as both T

_{C}and T

_{E}for CCs. However, T

_{E}for PTs is more akin to T

_{C}for CCs as both are business tax rates on net income, whereas the T

_{E}for CCs is the rate on equity payouts in the form of dividends and capital gains. For our tests, we used a 0.03 change in tax rates for each of the fifteen increasing P choices where each P choice corresponds to one of the fifteen Moody’s ratings used by Damodaran (2020). As described below, the use of 0.03 in conjunction with the setting of unlevered tax rates achieved effective levered tax rates that we expected to occur at the optimal debt-to-firm value ratio (ODV).

_{E}is 0.396), an effective T

_{E}for a pre-TCJA tax environment can be estimated at of 0.32. However, earlier sources (SBA 2009; National Federation of Independent Business 2013) suggest an effective T

_{E}that would be below 0.32 with variations based on the pre-TCJA years examined. Given this information, we selected an effective T

_{E}of 0.30 at ODV for pre-TCJA tests. Given this pre-TCJA estimate, a TCJA approximation for an effective T

_{E}at ODV would be about 0.28 due to the slight fall in personal tax rates caused by TCJA. Because we begin with an unlevered firm and allow tax rates to change when debt increases as described by Hull (2014), our pre-TCJA tests start with an unlevered T

_{E}of 0.35. This starting point enables us to achieve as an effective levered T

_{E}that is near 0.30 at ODV for pre-TCJA tests. For TCJA tests, we started with an unlevered tax rate of 0.33 to attain an effective levered T

_{E}that approaches 0.28 at ODV.

_{C}ranges from about 0.24 to 0.34 for a ten-year pre-TCJA period from 2001 through 2010. This suggests an effective T

_{C}of about 0.29 for the pre-TCJA tax environment. To achieve this rate, we set an unlevered T

_{C}at its maximum statutory rate of 0.35. The maximum corporate tax rate (T

_{C}) under TCJA is 0.21, which is also its minimum since 0.21 is a flat rate. Given that the estimated effective T

_{C}of 0.29 is 0.06 under the pre-TCJA maximum T

_{C}of 0.35, a value with the same drop of 0.06 in a TCJA environment would be 0.21 − 0.06 = 0.15. However, a proportional fall would be (0.29/0.35)0.21 = 0.174. Considering that T

_{C}is a flat rate under TCJA and tax credits and deductions may be more difficult to attain under TCJA, we considered an effective T

_{C}near 0.18 at ODV to be more reasonable and could be achieved with an unlevered T

_{C}of 0.21. As noted by Hull (2020) for his CC study under TCJA, an effective T

_{C}of 0.18 is consistent with the estimate of 0.18 given by the Penn Wharton Budget Model (2017) under TCJA. As described next, our tax rates on debt and CC equity income under TCJA were also consistent with Hull (2020).

_{D}) for both PTs and CCs where T

_{D}has a pre-TCJA maximum of 0.396 and a TCJA maximum of 0.37. If debt is held longer than three years, any capital gains is taxed at a lower capital gains rate where the typical maximum T

_{D}is 0.2. However, as noted by Hull (2020), wealthier investors often avoid paying taxes on for-profit debt by investing in nontaxable bonds because the taxable equivalent yield is higher. Given the above, we set T

_{D}at 0.22 as a reasonable effective tax rate for a typical PT or CC debt owner under a pre-TCJA tax environment. Under TCJA, we expect a slightly smaller T

_{D}and so set an effective T

_{D}at 0.21 as does Hull (2020). Unlike T

_{C}and T

_{E}that fall with leverage, T

_{D}rises with leverage. Thus, to achieves effective T

_{D}values near 0.22 and 0.21 at ODV, we set unlevered T

_{D}values at 0.19 and 0.18, respectively, for pre-TCJA and TCJA tests.

_{E}is typically only 0.2 and not the much higher personal statutory maximum for PTs. As mentioned by Hull (2020), investors have the ability to defer capital gains so that, through charitable contributions and inheritance, T

_{E}can be zero. For this reason, we advocate an effective personal tax rate (T

_{E}) on dividends and capital gains of about 0.14 for CC owners. To achieve an effective T

_{E}near 0.14 at ODV, the unlevered T

_{E}is set at 0.165. This rate holds for both pre-TCJA and TCJA tax environments since TCJA did not change the tax rates governing dividends and capital gains on equity.

#### 2.3.3. Effective Tax Rates for Robust Tests

_{E}may be small. Regardless, if such a provision is an important factor, then our estimate of T

_{E}for PTs is too high. In addition, PTs can buy shares in publicly traded companies or achieve capital gains on assets. If so, they pay taxes on qualified dividends and capital gains at the typical maximum T

_{E}of 0.2. Besides the above arguments for a lower effective tax rate for PTs, Hull and Price (2015) cite studies suggesting lower effective tax rates for both PTs and CCs. In light of the above factoids, we tested a set of lower effective tax rates on business income for PTs and CCs. Below we describe these effective tax rates.

_{E}, we performed robust tests where we seek a levered T

_{E}that is about 0.04 below that of our main PT tests. Similarly, we sought a levered T

_{C}that, on average, is about 0.04 below our main CC tests. To achieve the PT goal, we set the unlevered T

_{E}so it is 0.05 below that of our main PT tests. This serves to yield effective T

_{E}values for PTs near 0.26 and 0.24 for our respectively pre-TCJA and TCJA tests and thus 0.04 below the corresponding values of 0.30 and 0.28 for our main PT tests. Just like T

_{E}, disagreements can also be found for T

_{C}, as it also can vary from year to year with values often lower than what we use in our main tests. In response, our robust tests, that use a lower T

_{E}for PTs, also utilized a lower T

_{C}for CCs. While we lowered unlevered T

_{E}by the same amount (namely, 0.05) for both pre-TCJA and TCJA tests for PTs, such is not the case for CCs where the TCJA fall for T

_{C}is much greater than found for T

_{E}. Thus, we found it more appropriate to lower the unlevered T

_{C}for pre-TCJA tests by 0.07 and the unlevered T

_{C}for TCJA tests by only 0.03. The average of these values of 0.07 and 0.03 is 0.05. This average is like the PT tests where the unlevered T

_{E}is set 0.05 below for all of our main tests. Setting the unlevered T

_{C}(as just described) yields effective levered T

_{C}values at ODV that are 0.23 and 0.15 for respective pre-TCJA and TCJA tests. These levered T

_{C}values are about 0.06 and 0.02 below those for respective pre-TCJA and TCJA tests with an average of 0.04.

## 3. Methodology

#### 3.1. Capital Structure Models

#### 3.2. Identifying an Adequate Capital Structure Model

_{L}). This model focuses on the corporate tax shield when debt is issued to retire unlevered equity. The MM model claims the gain to leverage (G

_{L}) is the corporate tax rate (T

_{C}) time debt (D). Thus, this model is aimed at CCs. While this model allows for a perpetuity debt tax shield of T

_{C}(D), its G

_{L}equation ignores financial distress costs such as captured by increasing costs of borrowing as more and more debt is issued. To illustrate, G

_{L}is firm value (V

_{L}) minus unlevered firm value (V

_{U}) and this definition for G

_{L}includes variables that have both the unlevered cost of equity and levered cost of equity, which are not found in the MM equation of G

_{L}= T

_{C}(D). The MM equation for CCs also says nothing about growth, implies that a company can issue unrestricted amounts of debt, and ignores personal taxes paid on distributions.

_{L}= (1 − α)D where α captures the effects of personal and corporate taxes. However, like MM, the Miller model ignores financial distress effects, does not contain the costs of unlevered and levered equity, focuses on CCs, and can imply that a firm issues unlimited amounts of debt. In regard to the latter, the relation of α < 1 (which is the likely scenario for most ownership forms with reasonable leverage ratios) leads to the MM conclusion that firms attempt to issue unlimited amounts of debt. If α = 0, then G

_{L}= 0. If α > 1, then G

_{L}< 0. All of these outcomes for G

_{L}indicate that issuing debt does not produce an interior ODV. As such the Miller model does not reflect either conventional beliefs or observed managerial behavior such as managers targeting credit ratings (Kisgen 2009) to attain an interior ODV.

_{L}values as these models are not known for compact G

_{L}equations with measurable inputs. For example, how does one quantify all of the agency costs and benefits given by the agency theory literature? While a pecking order model is straightforward in terms of the order of financing preferences, it lacks comprehensive directives on how to measure the costs involved in the preferences for financing. For example, how do we accurately measure asymmetric information costs (which is a pecking order reason for not issuing external equity)? The pecking order model also does not address the cost of using internal equity (or retained earnings) for growth when compared to external equity where internal equity can only be used after it has been taxed. Whereas external equity is spared the potentially large costs associated with a maximum statutory business tax rate, it does experience equity issuance costs when issued for any purpose including growth. However, issuance costs represent a much smaller percentage of the proceeds from the issue compared to the tax rate paid by a typical business on retained earnings (RE) before it can be used for growth. In conclusion, agency and pecking order models do not offer formulations where all of its inputs are measurable inputs. It follows that these models are hard pressed to pinpoint preciseness in outputs for the three categories of debt choice, valuation and leverage gain where such preciseness is needed for correct capital structure decision-making.

_{L}equations that are consistent with trade-off arguments. Of practical importance, the CSM’s components use measurable variables, as we illustrate in our appendices. Of further significance, the CSM’s components are equipped to handle positive and negative effects. This is because these components contain not only positive agency-tax effects but also negative agency-financial distress effects. At some point, as the debt level becomes too high, an overall net negative effect occurs so that the CSM equations generate concave relations between firm value and leverage. These concave relations, that are needed to create an interior ODV, are graphically shown in Section 5.

_{U}and g

_{L}). The g

_{U}and g

_{L}equations for PTs with corresponding nongrowth and growth constraints was developed by Hull (2019), who also overviews the corresponding g

_{U}and g

_{L}equations for CCs. The g

_{U}and g

_{L}equations for CCs were first given by Hull (2010) with g

_{L}updated by Hull (2018) who added nongrowth and growth constraints for CCs. These constraints are violated when debt become too large. For example, with the plowback ratio (PBR) set (which implies the payout ratio with equity distributions is also set), it becomes more difficult for finite cash flows to cover interest payment as debt increases. With too much debt, this produces a situation where a firm is short of cash flows. Since debt owners have first claims, the firm does not have enough cash flows to meet its plowback and payout goals. As can be gleamed from comparing the nongrowth and growth outputs in later appendices, the growth constraint is violated at lower leverage ratios compared to the nongrowth constraint. This occurs because RE is zero with nongrowth and so more debt can be issued. Consequently, the growth constraint is also referred to as the retained earnings (RE) constraint as the CSM focuses on RE as the source of growth.

## 4. Pass-Through Variables, Procedures, Computations, and Applications

#### 4.1. Procedure to Match Credit Spreads, Ratings, and Costs of Borrowing to Debt-to-Firm Value Ratios

_{U}) retired with debt (D). The increase in borrowing costs as leverage increases is a key to discovering the maximum firm value that identifies ODV. In this table, we follow the procedure given by Hull (2020) for his nongrowth and growth CC tests when using interest coverage ratios (ICRs), credit spreads, and Moody’s ratings as supplied by Damodaran (2019) for 2018. Table 1 provides value when using Hull’s procedure for a nongrowth pass-through (PT) test under TCJA using ICRs, spreads, and ratings supplied by Damodaran (2020) for 2019. The ICRs and credit ratings given by Damodaran do not change from 2018 to 2019 but the credit spreads (that are matched to the ICRs and ratings) do change. The same procedure described in Table 1 for a nongrowth PT test under TCJA is used to get costs of borrowing matched to P choices for all of this paper’s other tests (that include CC, growth, and pre-TCJA tests). The values presented and derived in Table 1 are needed when applying the CSM equations, as illustrated in the three appendices.

_{BT}) are the same for each P choice. Similarly (and as can be seen in Table 1), because costs of borrowing (like tax rates) are identical for each P choice, debt (D) is also same for each P choice for all PT nongrowth and growth tests. While I and D are identical for each P choice for PT growth and nongrowth tests, values for P choices and DVs are different for nongrowth and growth tests because they are a function of equity value and the growth equity value is different from the nongrowth equity value for each test. The latter difference is detailed in Appendix A, while illustrative computations for P choices and DVs (along with other variables) are detailed in Appendix B and Appendix C.

#### 4.2. Procedures to Determine Optimal Outputs

#### 4.2.1. Procedure with No Growth

_{L}equation is an easy task as one and only one maximum firm value (max V

_{L}) was achieved, as illustrated later in Section 5. For nongrowth tests, the simple procedure involves examining all feasible V

_{L}outputs generated by the nongrowth G

_{L}equation and choosing the largest V

_{L}where V

_{L}= G

_{L}+ unlevered firm value (E

_{U}). The largest value V

_{L}is referred to as the maximum V

_{L}(max V

_{L}) where max V

_{L}= max G

_{L}+ E

_{U}. Thus, either max V

_{L}or max G

_{L}can identify all other optimal nongrowth outputs, including ODV and OCR where the nongrowth OCR for PTs is Moody’s A3 (for CCs, the nongrowth OCR is Moody’s Baa2). Optimal outputs for a nongrowth PT are computed in Appendix B where we can see in that the largest V

_{L}occurs at A3.

#### 4.2.2. Procedure with Unrestricted Growth

_{L}equation is not simple especially if we have many possible growth rates to test where each growth rate is represented in the CSM by the levered equity growth rates (g

_{L}). If we do not restrict g

_{L}, then we can find (through trial and error) the plowback ratio (PBR) that yields the largest V

_{L}. However, this procedure leads to extremely high g

_{L}values as V

_{L}increases and these g

_{L}values are unsustainable for long periods based on historical growth in US GDP. Thus, we cannot claim that the largest V

_{L}is max V

_{L}. Regardless, this unrestricted growth procedure is informative because as we initially began to increase the PBR and achieve high (but still sustainable) g

_{L}values, we found the largest V

_{L}consistently occurs at the same nongrowth OCR. As PBR and g

_{L}continue to increase, the largest V

_{L}will occur at a lower quality rating that is often just a notch below in quality from the nongrowth OCR. However, this lower quality rating can only be achieved with growth rates that are historically very difficult to sustain for long-run time frames. Thus, at least for this paper’s tests, we concluded that max V

_{L}for growth occurs at the same OCR, as found with our nongrowth test.

_{L}values that are even more unsustainable, we find that the growth constraint begins to set in at lower P choices or, equivalently, set in at higher quality ratings. Eventually, this limits the feasible P choices so that only high quality credit ratings become feasible and V

_{L}can fall with high ratings but for some tests can increase and even occur with growth rates that are not much greater than the long-run historical growth rate. This is because g

_{L}is a levered rate that increases with greater debt choices. This means that g

_{L}can fall as greater debt choices becomes unfeasible because of violation of the growth constraint. Thus, it becomes possible to find a larger V

_{L}at a higher quality rating with a g

_{L}that is historically sustainable. However, we know an average firm cannot attain higher quality ratings and so we have to question these outcomes, at least for a typical firm. It is during these tests with increasing growth (accompanied by violations of the CSM’s growth constraints) that computations become unstable for any credit rating that is accompanied by high levels of debt where the growth constraint can be violated. This instability also reflects the reality, as first noted by Hull (2010), that the G

_{L}equation requires an iterative procedure because g

_{L}and G

_{L}are interdependent.

_{U}) becomes feasible because the growth constraint is violated for every credit rating. When a levered firm becomes unlevered, then V

_{L}turns into E

_{U}and g

_{L}becomes an unlevered growth rate (g

_{U}) where interest (I) can no longer compete with RE for use of the operational cash flows. At this point, it is possible for E

_{U}to compete with the highest V

_{L}and do so while maintaining a growth rate that is not abnormally high. However, as PBR continues to increase for this unlevered situation, g

_{U}will increase along with a rising E

_{U}. Like the larger V

_{L}values with credit ratings of higher quality than the nongrowth OCR, the larger E

_{U}values can only be reached with a historically unsustainable growth rate that is unrealistically high, especially for a perpetuity model like the CSM.

#### 4.2.3. Procedure with Restricted Growth

_{L}. We circumvented such complications and impracticalities by conducting tests where we restrict growth rates to a finite number of reasonable rates. For our first restricted growth test, we used the long-run sustainable growth rate of 3.12%, which was first described in Section 4.1. For this rate for both pre-TCJA and TCJA tests, we examined all feasible credit rating with g

_{L}= 3.12% to see which rating generates the greatest firm value. We can stop here for pre-TCJA tests but not for TCJA tests because a higher growth rate is projected under TCJA. For reasons given in Section 4.1, we used g

_{L}= 3.90% and g

_{L}= 4.50% for two additional growth tests under TCJA. As before, we then proceeded to test all feasible credit ratings in conjunction with these two growth rate before making a decision on max V

_{L}that identifies the growth OCR. We now describe the outcomes for the above tests.

_{L}= 3.12% for our first growth PT test under both pre-TCJA and TCJA. To get g

_{L}= 3.12% for each credit rating being tested, we set PBR through trial and error until g

_{L}= 3.12% occurs for a P choice that corresponds to the rating being tested. For both pre-TCJA and TCJA tests for PTs, we found that the nongrowth OCR of A3 is also the growth OCR. The optimal outputs for the 3.12% growth PT test under TCJA are computed in Appendix C. From this appendix, we can see that the largest V

_{L}occurs at A3. Thus, this largest V

_{L}is max V

_{L}and it identifies optimal outputs including A3 as the OCR.

_{L}= 3.12% (for both pre-TCJA and TCJA tests) produce two differences from PTs. First, the nongrowth tests for CCs under pre-TCJA and TCJA yield a lower medium grade rating of Moody’s Baa2 as the nongrowth OCR. Second, we can find higher V

_{L}values for ratings of higher quality than Baa2 when g

_{L}= 3.12% is tested. Thus, unlike PTs, we cannot confirm for CCs that the same OCR occurs for both nongrowth and growth of 3.12% for pre-TCJA and TCJA tests. However, evidence indicates we should be hesitant in claiming a typical CC can attain a rating greater than Baa2 while achieving g

_{L}= 3.12%. For example, as indicated by Morningstar (2019), high quality ratings are becoming rarer. Thus, when using a growth rate of 3.12%, we concluded that both a nongrowth and growth CC will typically achieve its OCR at a rating of Baa2. Even though we concluded that a typical growth CC will achieve its max V

_{L}at Baa2, we would also advocate that an above average CCs can attain higher quality ratings at g

_{L}= 3.12% and therefore have higher max V

_{L}values. Regardless, the increase in max V

_{L}is small even if we claim the higher max V

_{L}is achievable for some CCs. For example, for the TCJA test, we found only a 2.66% increase in max V

_{L}if a CC can attain an OCR of A3 with g

_{L}= 3.12% instead of an OCR of Baa2 with g

_{L}= 3.12%. While max V

_{L}only increases by 2.66% for this CC test, its ODV would falls by 19.22%. This indicates that our computations for ODVs can be sensitive to the OCR that is achieved.

_{L}= 3.90% and g

_{L}= 4.50% under a TCJA tax environment. As described previously, these two annual growth rates under TCJA are consistent with the sources cited by the Tax Policy Center (2018). An annual growth rate of 3.90% is also consistent with periods of 20 to 25 years that occur within the past 70 years. To find an average annual growth of 4.50% for a period longer than 10 years, we would have to go back 85 years and include years of extreme growth associated with the 10-year period from 1934 through 1943 where the average annual growth in GDP was 10.49%.

_{L}= 3.90% and g

_{L}= 4.50% tests, CC and PT results are like those just outlined for g

_{L}= 3.12% tests except for the PT test for g

_{L}= 4.50%. For this latter PT test, the nongrowth OCR of A3 is not attainable due to violation of the growth constraint at A3. To achieve g

_{L}= 4.50% for PTs, we had to drop the nongrowth OCR a notch from A3 to Baa2. For higher growth rate tests for both PTs and CCs, we can occasionally find greater V

_{L}values for credit ratings of higher quality than the nongrowth OCR. This is especially true for the g

_{L}= 3.90% test. For these situations (where greater V

_{L}values occur for credit ratings of higher quality), we once again concluded that the strongest firms can attain the same high growth with ratings of higher quality than the quality of the nongrowth OCR. For example, if a PT can achieve a higher quality rating of A2 with a growth rate of 3.90%, max V

_{L}can increase but by only 2.62%. However, we maintain that an average firm (which is what this study focuses on) is not likely to achieve an OCR of greater quality than the nongrowth OCR. To maintain the same high growth rate of 3.90% or 4.50% at a higher credit rating, PBR must increase more and more as the credit rating increases in quality and so more internal equity funds would be needed to maintain the same rate of growth. This could prove very difficult even in the short-run without the infusion of cash from a new external security issue that supplies funds for growth.

_{L}= 4.50% tests) and a typical CC consistently attains a Moody’s credit rating of Baa2 at ODV. The lower credit rating for a CC, compared to a PT, is congruent with the notion that larger firms are less risky and investors would be willing to buy more debt even when lower quality ratings occur. As shown later in Section 6, we find this notion to be valid as the average ODV is 0.2377 for PT tests compared to 0.4132 for CC tests. As also shown later, some of our findings are subject to change when we tested spreads for 2017 and 2018 but not so much when we tested lower effective business tax rates for PTs and CCs.

#### 4.3. Introductory Variables and Computations Used by Capital Structure Model (CSM)

_{L}is 3.12%. The computations use the Capital Structure Model (CSM) of Hull (2019) for PTs with tax rates under TCJA. The computations featured in Appendix A are as follows.

_{1}) is the Miller (1977) alpha that was first derived by Farrar and Selwyn (1967) and updated by Hull to account for changes in tax rates as leverage changes. The second coefficient (α

_{2}) is the Hull alpha. Because the same tax rates occur with each P choice for all PT tests, values for α

_{1}and α

_{2}are the same for all PT tests and are respectively found in the 1st and 2nd components of the G

_{L}equations for PTs. The latter also holds for CC tests, albeit alpha values for CCs are different because they have slightly different formulas due to the extra layer of corporate taxes. Regardless, α

_{1}and α

_{2}values increase with leverage for both PT and CC tests. Appendix A computes α

_{1}and α

_{2}when the PT achieves its OCR at Moody’s A3 under TCJA. As seen in this appendix, these alpha values are α

_{1}= 0.905586 and α

_{2}= 1.069579. For a nongrowth G

_{L}equation, an increasing α

_{1}makes the 1st component more positive while an increasing α

_{2}makes the 2nd component more negative. For a growth G

_{L}equation, an increasing α

_{1}makes the 1st component more positive but then reverses its effect, while an increasing α

_{2}lessens its negative effect on the 2nd component and then makes this 2nd component more positive until the growth constraint is violated.

_{U}) under TCJA, Appendix A offers a PT nongrowth example and a PT growth example for g

_{L}= 3.12%. The starting point to compute E

_{U}is $1,000,000 in before-tax cash flow1. For the nongrowth example where PBR = 0 and the unlevered personal equity tax rate (T

_{E}

_{1}) = 0.33, the nongrowth E

_{U}was computed as $9,922,251. For the growth example where the optimal PBR is 0.3435 at ODV, the growth E

_{U}was shown to be $10,030,170 and thus growth adds $107,919. While our pre-TCJA test where T

_{E}

_{1}= 0.35 is not shown in Appendix A, we obtained a nongrowth E

_{U}of $9,626,064 and a growth E

_{U}of $9,638,148 with a PBR of 0.3515 and so only $12,084 was now added from growth. Thus, an increase of T

_{E}

_{1}from 0.33 to 0.35 made the difference in nongrowth E

_{U}and growth E

_{U}fall from $107,919 to $12,084. If we continued to increase T

_{E}

_{1}, then the nongrowth E

_{U}would become greater than a growth E

_{U}and value would be subtracted instead of added with growth. This reflects the fact that RE used for growth becomes more expensive as it is taxed at higher tax rates.

_{U}will be greater than the growth E

_{U}? The answer lies in the discussion by Hull (2010, 2018) of the relation between taxes on retained earnings (RE) and PBR. Hull proves the nongrowth E

_{U}is greater than the growth E

_{U}when the cost of internal equity financing is greater than PBR. For Hull, the cost of using internal equity financing involves the business taxes paid on RE before it can be used to finance growth. As illustrated in the prior paragraph for the TCJA test, when the cost of unlevered T

_{E}

_{1}is 0.33 and thus lower than PBR of 0.3435, an unlevered firm added $107,919 in value through growth. However, the nongrowth E

_{U}and growth E

_{U}were similar for a pre-TCJA environment with only $12,084 in value added when the cost of unlevered T

_{E}

_{1}increases from 0.33 to 0.35 (with the PBR increasing from 0.3435 to 0.3515, which is about 3/8 as much of an increase compared to T

_{E1}). As will be seen later in Section 6 for the pre-TCJA tests for CCs, the nongrowth max V

_{L}of $10.032 M is greater than the growth max V

_{L}of $9.476 M when g

_{L}= 3.12%. This reflects the fact that the unlevered nongrowth E

_{U}is greater than the unlevered growth E

_{U}where RE is taxed at the pre-TCJA unlevered corporate tax rate (T

_{E}

_{2}) of 0.35, which is greater than the PBR of 0.3366. In conclusion, PBR must be greater than the business level tax rate on RE if growth to enhance nongrowth E

_{U}. If it cannot, then the nongrowth max V

_{L}is more likely to be greater than the growth max V

_{L}. This is why TCJA is important as expectations are that growth will increase beyond 3.12%, causing PBR to increase so that it will be greater than the unlevered tax rate (T

_{E}

_{1}for PTs and T

_{E}

_{2}for CCs) and make growth more affordable for firms.

#### 4.4. Pass-Through Applications: Nongrowth and Growth under TCJA

_{L}equation of Hull (2019). This appendix includes a table that reports values for variables for eleven of the sixteen P choices where the sixteen P choices consist of an unlevered P choice of zero and fifteen interior P choices corresponding to the fifteen interest coverage ratios (ICRs) with matching ratings and spreads supplied by Damodaran (2020) for the year 2019. The bold print column in the table in Appendix B with the OCR of A3 provides optimal values ranging from P = 0.2582 in the top row to ODV = 0.2409 in the bottom row.

_{L}values first occur when the nongrowth PT goes from a non-investment grade credit rating (Moody’s Ba1) to a lower quality non-investment rating (Moody’s Ba2). The next lower quality rating (Moody’s B1) is a highly speculative credit rating. An even lower quality rating of B2 is in the last column. While not shown in Appendix B, if we were to issue enough debt to achieve a Moody’s Caa rating (which is a rating that indicates extreme speculation bordering on default), the PT nongrowth constraint would be violated. This constraint is breeched at high debt levels when the firm no longer has the cash flows to cover interest (I) with part of the reason being lower (or even negative) G

_{L}values. Whereas positive G

_{L}values can provide funds to pay I, negative G

_{L}values lower funds that could otherwise be used to pay I.

_{L}of $10.633M (M = million) at its ODV of 0.2409. For example, the max %∆E

_{U}was computed as 7.16% at this ODV. This percentage of 7.16% indicates leverage adds, on average, 7.16 cents to unlevered firm value (E

_{U}) for every dollar of debt issued to retire E

_{U}when a nongrowth PT is at this ODV. The net benefit from leverage (NB) was computed as 27.74%, indicating that each dollar of debt, on average, adds 27.74 cents to G

_{L}when the PT is at its ODV. A greater value for NB indicates greater efficiency per dollar of debt issued.

_{L}) of 3.12%. This appendix uses the PT growth G

_{L}equation of Hull (2019). The growth OCR (achieved with a PBR of 0.3435), like the nongrowth OCR seen in Appendix B, has a Moody’s upper medium rating of A3. The bold print column of the table that accompanies Appendix C provides optimal values ranging from P = 0.2554 in the top row to ODV = 0.2381 in the bottom row. The last two columns are in gray-shade to signify that their values are unfeasible, as the growth constraint is violated once we achieve a Moody’s credit rating of B1, which is a highly speculative rating.

_{U}was computed as 7.25%, indicating the unlevered firm value increases by 7.25% when the optimal amount of debt is issued. This is a bit above the 7.16% achieved for the nongrowth situation. While we can note other differences in nongrowth versus growth values when comparing the accompanying tables in Appendix B and Appendix C, these differences are also small. For example, we found slightly larger values with growth for the outputs of max G

_{L}, max V

_{L}, and NB when compared to nongrowth values. The value for ODV of 0.2381 with growth was slightly lower than the ODV of 0.2409 for nongrowth. This slightly lower ODV is explained by the greater equity value for growth as the fifteen debt values are the same for either a nongrowth test or a growth test. They are the same because debt values are derived from interest coverage ratios (ICRs) and the same ICR values are used for all PT tests (as described in Section 4.1). Finally, while a growth rate of 3.12% does not produce noteworthy differences with nongrowth outputs, Section 6 will show that this is not the case for higher growth rates of 3.90% and 4.50%.

## 5. Pass-Through Results in Graphical Form

_{L}) and its two components against debt-to-firm value ratios (DVs) under TCJA when there is nongrowth. Figure 2 repeats Figure 1 but replaces the nongrowth values with growth values using the historical growth rate of 3.12%. Figure 3 plots firm value (V

_{L}) against credit ratings for the TCJA nongrowth test and the three TCJA growth tests. As first described in Section 1, these three growth tests involve growth rates of 3.12%, 3.90%, and 4.50%. Figure 3 only considers the feasible P choices for growth tests causing the three growth trajectories to have different plot points. Unfeasible choices are not possible due to violation of the growth constraint. Whereas growth trajectories only consider feasible points, the nongrowth trajectory is truncated in that all of its feasible points are not shown.

#### 5.1. Gain to Leverage versus Debt-to-Firm Value Ratio

_{L}equation of Hull (2019) that is illustrated in Appendix B. This trajectory ends with DV = 0.3691, which corresponds to a Moody’s credit rating of B1. While not shown, the last feasible DV of 0.5201 corresponds to a rating of B3. As was seen in Section 4, the next rating below B3 in quality is Caa and this latter rating is where the nongrowth constraint is violated. The 1st component given by the top trajectory (dotted line) has an upward trend until the first downward bump occurs at DV = 0.3037. The bottom trajectory (dashed line), representing the 2nd component, is decreasing with the drop-off becoming steeper as more debt is issued. These trajectories are consistent with the notion that the 1st component captures the positive leverage effects from tax and agency benefits related to debt while the 2nd component capture the negative leverage effects from financial distress and agency costs.

_{L}nongrowth trajectory (given by the solid line) of Figure 1 is concave in shape and peaks at the ODV of 0.2409 where the OCR is Moody’s A3 rating. There is asymmetry in the G

_{L}nongrowth trajectory where the fall off is greater after the maximum G

_{L}(max G

_{L}) of $0.710M (M = millions) is reached where $0.710M is per $1,000,000 in before-tax cash flows. G

_{L}first becomes negative for a rating of Ba2 with a DV of 0.3353, which is before it enters the highly speculative rating of B1 where DV = 0.3691. From here the negativity in G

_{L}continues as debt choices rise until the nongrowth constraint is violated, at which point all outputs are unattainable. As noted above, the last feasible DV (before the constraint violation occurs) is 0.5201, which corresponds to a Moody’s rating of B3. All ratings of lower quality than B3 are unfeasible and cannot support the debt choice associated with such a low quality rating.

_{L}equation of Hull (2019). Results for this equation are illustrated in Appendix C. Figure 2 contains only feasible plot points because values after DV = 0.3152 are unattainable as the growth constraint is violated. A DV of 0.3152 corresponds to a Moody’s rating of Ba2, which is the second non-investment or speculative rating given by Damodaran (2020). The 1st component in Figure 2 has a trajectory (dotted line) that is concave in shape like the 1st component in Figure 1 except it has a greater drop-off after its peak is reached at DV = 0.1936. The 2nd component in Figure 2 has an upward trajectory (dashed line) that is concave upward but does not exhibit full concavity. This trajectory contrasts with the downward trajectory of the 2nd component in Figure 1. Regardless, this trajectory’s first three plot points are negative and its later positive plot points are offset by negative plot points in the trajectory for the 1st component.

_{L}growth trajectory (solid line) in Figure 2 is concave in shape like the G

_{L}nongrowth trajectory in Figure 1. The G

_{L}growth trajectory peaks at the ODV of 0.2381 where max G

_{L}= $0.728 M. While not shown, similar trajectories for G

_{L}and its two components occur if we increased the annual growth rate to g

_{L}= 3.90% or g

_{L}= 4.50%. However, as will be seen in Figure 3, full concavity becomes more difficult to achieve as the growth rate continues to increase.

_{L}values were similar to growth G

_{L}values until we reached the optimal P choices, at which point growth G

_{L}values became greater and continued to increase relative to nongrowth values until the growth constraint was violated.

#### 5.2. Pass-Through Firm Value versus Credit Ratings

_{L}) versus credit ratings for the nongrowth PT tests under TCJA and three growth PT tests under TCJA. This figure differs from the prior two figures by replacing G

_{L}with V

_{L}along the vertical axis and DV with Moody credit ratings, as used by Damodaran (2020), along the horizontal axis. For Figure 3, we only include credit ratings up to Ba2. As seen in this figure, the two trajectories for the two greatest growth rates (3.90% and 4.15%) cannot reach this rating of Ba2, as the growth constraint is violated at quality ratings greater than Ba2. For the nongrowth PT trajectory (dotted line at bottom), violation of the nongrowth constraint occurs at the extremely speculative rating of Caa, which is well below Ba2 in quality as was shown in Section 4. For the growth trajectory (small dashed line) when g

_{L}= 3.12%, violation of the growth constraint first occurs with a rating of B1, which is a rating just below Ba2 in quality. For the growth trajectory (solid line) when g

_{L}= 3.90%, violation of the growth constraint occurs at a rating of Ba2 and so this trajectory ends at a rating of Ba1. For the growth trajectory (large dashed line at top) when g

_{L}= 4.50%, violation starts with Ba1 so the trajectory stops at Baa2. Given this pattern, we concluded that greater growth leads to violations at higher quality credit ratings.

_{L}= $10.633 M at the OCR of A3. For the dashed line trajectory where g

_{L}= 3.12%, the plowback ratio (PBR) of 0.3435 yields g

_{L}= 3.12% for an OCR of A3 with max V

_{L}= $10.758 M. The latter max V

_{L}is a $0.125 M larger than the nongrowth max V

_{L}of $10.633 M in the dotted line trajectory. In percentage form, the increase is only 1.18%. The small difference of $0.125 M is not caused by the choice of credit ratings. For example (and as alluded to in our discussion in Section 4.2), if we set PBR so that g

_{L}= 3.12% for both lower and higher quality credit ratings, all V

_{L}values are lower than the growth max V

_{L}of $10.758 M that occurs at A3. Thus, a lower or higher quality credit rating does not necessarily lead to increased firm value, even when growth is the same. Thus, at least for this growth test, targeting the correct credit rating as OCR is important.

_{L}is $11.506 M when g

_{L}= 3.90% at a rating of A3 with a PBR of 0.3934. This V

_{L}of $1.506 M is an 8.22% increase over nongrowth max V

_{L}of $10.633 M. Furthermore, the V

_{L}of $11.506 M for g

_{L}= 3.90% is 6.95% greater than the max V

_{L}of $10.758 M when g

_{L}= 3.12%. However, this solid line trajectory reveals that the greatest V

_{L}is not $11.506 M that occurs at A3, but $11.527 M where the latter is attained at a rating of Baa2. Nevertheless, if the largest sustainable growth rate is 3.90%, we could not consider $11.527 M as the max V

_{L}because it is attained with g

_{L}= 4.33%. Thus, we concluded that $11.506 M is still the likely candidate for max V

_{L}if 3.90%. Below we explore this candidacy.

_{L}= 3.90% occurs for a lower quality rating of Baa2, we found V

_{L}= $11.056 M, which is below $11.506 M when g

_{L}= 3.90% for a rating of A3. If we set g

_{L}= 3.90% for a higher quality rating of A2, we found V

_{L}= $11.588 M, which is above $11.506 M when g

_{L}= 3.90% for a rating of A3. While this test yields a larger V

_{L}, other tests of higher quality ratings often yield either lower V

_{L}values or unfeasible outcomes where the latter occurs because higher quality ratings can lead to the growth constraint being violated. Regardless, if a PT manager can achieve growth of 3.90% at a higher quality rating, then greater firm value can result and determining max V

_{L}and OCR becomes a question as to what is the highest quality credit rating that can be attained by a PT when g

_{L}= 3.90%. Achieving a higher V

_{L}for a higher quality credit rating than A3 for g

_{L}= 3.90% differs from what we just saw for the lower growth rate of g

_{L}= 3.12% where achieving 3.12% for a higher quality rating than A3 did not increase max V

_{L}. Given the difficulty of attaining and maintaining higher quality credit ratings, we would argue that a typical PT has an OCR of A3 if it can attain g

_{L}= 3.90%. However, a stronger PT should be able to achieve an OCR of higher quality than A3 when g

_{L}= 3.90%, while a weaker PT would have to settle for an OCR of lower quality than A3.

_{L}= 4.50%. As described in Section 1, a rate of 4.50% is consistent with the high end of the growth spectrum cited by the Tax Policy Center (2018) under TCJA. For this trajectory, we use an OCR of Baa2 because an A3 rating with a growth rate of 4.50% is not feasible. This is because the growth constraint is violated when an A3 rating (or any rating of higher quality than Baa2) is used with g

_{L}= 4.50%. The difficulty in achieving the higher quality rating of A3 is consistent with what we found in the real world, where higher quality ratings are unachievable for an average firm. The V

_{L}of $11.756 M in the top trajectory for a Baa2 rating when g

_{L}= 4.50% is greater than the V

_{L}value of $11.506 M that occurs when g

_{L}= 3.90% at an A3 rating in the solid line trajectory. It can be pointed out that the V

_{L}value of $11.712 M for an A3 rating in the top trajectory is achieved with an annual growth rate of 4.06% and this value of $11.712M is also higher than $11.506 M achieved at g

_{L}= 3.90% with a rating of A3 in the solid line trajectory. As seen in the top trajectory, the firm values of $11.299 M and $11.498 M for credit ratings of A1 and A2 (with respective growth rates of 3.69% and 3.84%) are lower than $11.506 M achieved at g

_{L}= 3.90% with a rating of A3. When we tried to target a Ba1 or lower quality rating when g

_{L}= 4.50%, we found that V

_{L}values fell. From the 4.50% growth tests (just like the 3.90% growth test), we deduced that firm value can depend on the quality of the rating that is achievable for a higher growth rate projected under TCJA.

_{L}= 3.12%. As expected (and shown next in Section 6), pre-TCJA tests yield smaller max V

_{L}values than those for TCJA since the pre-TCJA cash flows are taxed at greater levels. Regardless, there are similarities in all pre-TCJA and TCJA trajectories in that we find concave relations between V

_{L}and credit ratings. The pre-TCJA nongrowth max V

_{L}for PTs is only $0.019 M less than its growth max V

_{L}. This pre-TCJA difference of $0.019 M is smaller than the difference of $0.125 M for the same TCJA comparison for PTs. Since pre-TCJA tax rates are higher for PTs than TCJA tax rates, we surmised that higher pre-TCJA business tax rates make it more difficult for growth V

_{L}values to compete with nongrowth V

_{L}values. This is because RE is used for growth only after it is taxed and a pre-TCJA environment has greater business level tax rates, making the use of RE more expensive. This will be better seen in the next section, when we present CC results where the pre-TCJA business level corporate tax rate is significantly higher than that for TCJA.

## 6. Results for Pass-Throughs and C Corps

#### 6.1. Optimal Outputs for the Three Categories of Debt Choice, Valuation, and Leverage Gain

_{E}= 0.3006 and T

_{D}= 0.2203 for the two pre-TCJA tests in the first two rows; T

_{E}= 0.2834 and T

_{D}= 0.2087 for the first three TCJA tests in the next three rows, and T

_{E}= 0.2749 and T

_{D}= 0.2149 for the last test (where g

_{L}= 4.50%). For CCs in Panel B, we have T

_{C}= 0.2915, T

_{E}= 0.1374, and T

_{D}= 0.2269 for the two pre-TCJA tests in the first two rows and T

_{C}= 0.1749, T

_{E}= 0.1374, and T

_{D}= 0.2149 for the four TCJA tests in the last four rows.

#### 6.1.1. Debt Choice Outputs

#### 6.1.2. Valuation Outputs

_{U}and maximum firm value (max V

_{L}) are $10.111 M and $10.902 M, respectively, in the last row of Panel A. These averages compare to those for CCs in the last row of Panel B where E

_{U}and max V

_{L}average $9.520 M and $11.120 M, respectively. While the PT average for E

_{U}is 6.21% higher than the CC average, the PT average for max V

_{L}is 1.95% lower than corresponding CC average. This same pattern is observed when replacing 2019 spreads with either 2017 spreads or 2018 spreads. Panel B of Table 2 reveals larger CC valuations occur for the TCJA tests where CCs receive more favorable tax treatment than PTs. While these larger CC valuations are predictable given the huge drop-off in the effective corporate tax rate under TCJA, what Table 2 offers is a detailed look at the numbers that reveal precisely how TCJA can increase pre-TCJA firm valuation and how increasing growth consistently increases firm value. Thus, the valuation outputs in Table 2 confirm the notion advanced by tax experts, such as those at the Tax Policy Center (2018), that lower taxes are necessary to prevent economic harm and increase business wealth.

_{L}values for our two pre-TCJA tests for PTs with those for CCs, we find CCs, on average, have a valuation disadvantage of 6.037%. For the eight TCJA tests (four for PTs and four for CCs), the valuation advantage now favors CCs as its average max V

_{L}is 5.727% greater than PTs. The net percentage gain for an average CC over an average PT caused by TCJA is 6.037% + 5.727% = 11.764%. This reveals that TCJA can put an average PT in a better position to increase its value by switching its ownership form to a CC. This conclusion is not a product of using spreads for 2019, as 11.764% increases to 13.525% and 13.716% if we use the respective spreads for 2017 and 2018. Together these results indicate that, ceteris paribus, the future will find more businesses claiming the CC ownership form.

#### 6.1.3. Leverage Gain Outputs

_{L}, max %∆E

_{U}and NB are $0.791 M, 7.81% and 30.48%, respectively. The PT averages for G

_{L}and max %∆E

_{U}are substantially less than those for CCs in Panel B where the values are $1.600 M for max G

_{L}and 17.20% for max %∆E

_{U}. This reveals that CCs get greater gains from leverage both in dollars and percentages. In terms of NB, PTs are now more similar to CCs by having an average of 30.48% compared to 33.05% for CCs. This indicates that PTs are only slightly less efficient than CCs by getting a bit smaller gain for every dollar of debt added. The average PT output of 7.81% for max %∆E

_{U}is like the prior empirical research (Graham 2000; Korteweg 2010; Van Binsbergen et al. 2010) where max %∆E

_{U}values range from 4% to 10%. However, the average CC output of 17.20% for max %∆E

_{U}is greater than prior research. For both ownership forms, we find an overall average of 12.50%, which is still high historically. As was just argued for our ODV results, we can also explain this difference in terms of the recent data from Damodaran (2020) with spreads for 2019. For example, using spreads for 2018, we found the overall average of 12.50% changes to 7.33%, which is consistent with the empirical research where the midpoint is 7%. We concluded that leverage gain outputs, like debt choice outputs, can be sensitive to the annual changes in the spreads used by Damodaran.

#### 6.2. Outputs With Lower Tax Rates

_{L}value. Of further importance, it is still advantageous for a PT ownership form to switch to a CC ownership form. However, the 11.764% advantage given earlier for CCs fell to 6.479%. Since our tests lowered PT and CC tax rates by about the same amount, we concluded that lowering effective tax rates in a uniform manner can make it less likely for a PT to switch to a CC.

_{L}values increase with lower tax rates. This is consistent with greater max V

_{L}values increasing with lower tax rates, since greater max G

_{L}values ceteris paribus lead to greater max V

_{L}values. We found that max %∆E

_{U}values fall for all tests. In respect to the overall average max %∆E

_{U}of 12.50% gleamed from Table 2, we find that 12.50% falls to 10.49% with lower effective tax rates. Similar to max G

_{L}and max %∆E

_{U}values, all NB values also decline.

#### 6.3. Policy Implications

#### 6.4. Future Research

## 7. Summary

_{L}) that is a function of interest payments and earnings retained for growth. Each CSM nongrowth and growth test receives the same before-tax cash flow with borrowing costs linked to credit ratings and spreads. For our first growth test, we use a g

_{L}of 3.12%, which is based on annual US real GDP growth data for a recent seventy-year horizon. Due to increased growth projected under TCJA, we also use annual growth rates of 3.90% (consistent with what tax experts expect to occur under TCJA) and 4.50% (consistent with the greatest growth expected to occur under TCJA). Below we summarize our main findings.

_{L}) under TCJA where tax rates are lower. While this is expected, we report precise dollar amounts detailing how much higher TCJA max V

_{L}values are compared to pre-TCJA max V

_{L}values. Except for increased growth under TCJA (where 3.12% increases to either 3.90% or 4.50%), nongrowth max V

_{L}values are similar to growth max V

_{L}values with the pre-TCJA nongrowth max V

_{L}for CCs actually slightly greater than its growth max V

_{L}. The latter shows that high pre-TCJA corporate tax rates make growth unaffordable for an average CC when growth is only 3.12%. However, if growth increases as anticipated under TCJA, we find noticeably greater max V

_{L}values. We show that the greater fall in tax rates for CCs under TCJA cause max V

_{L}values for CCs to be greater than those for PTs. This is opposite of that found for pre-TCJA tests and suggests a typical PT can, ceteris paribus, profit by becoming a CC under TCJA. In conclusion, we show that (i) firm valuation outputs increase under TCJA and this is especially true when growth increases beyond its pre-TCJA norm of 3.12% as projected by tax experts, and (ii) a typically PT can increase its value by switching to the CC ownership form.

_{L}) and the maximum percentage increase in unlevered equity (max %∆E

_{U}). In terms of the net benefit from leverage (NB), CCs outperform PTs for the two pre-TCJA tests and the nongrowth TCJA test. However, we discover NB values for PTs are greater than CCs for the three TCJA growth tests indicating that PTs attain a greater gain per dollar of debt issued under TCJA when growth is present. In conclusion, PTs have substantially lower values compared to CCs for max G

_{L}and max %∆E

_{U}but can have higher values for NB, with the latter occurring for the three TCJA growth tests.

## Funding

## Acknowledgments

## Conflicts of Interest

## Appendix A. Computations for Introductory Variables Used by CSM

_{1}and α

_{2}) where α

_{1}and α

_{2}capture the tax rate effect in the 1st and 2nd components of the CSM G

_{L}equation for PTs given by Hull (2019). These computations are the same for both nongrowth and growth of 3.12% since the optimal credit rating (OCR) of A3 is the same for both. The annual growth rate of 3.12% is consistent with the seventy-year growth rate in GDP as supplied by the US Bureau of Economic Analysis (2020). The personal tax rate on equity (T

_{E}) and the personal tax rate on debt (T

_{D}) are allowed to change in their expected directions as discussed in Section 2.3.2.

_{U}) used by the CSM. We begin by setting the unlevered equity beta (β

_{U}) at 0.79, which is consistent with data given by Damodaran (2020) for 2019 for the average firm. This beta is also consistent with Damodaran (2019) for his data for 2018. For our tests, a β

_{U}of 0.79 yields a levered beta (β

_{L}) near 1.00 when a firm is at ODV, which is consistent with the notion that an average firm would have a beta equal to the market beta (β

_{M}) of 1.00. Using the CAPM (as seen below), we can compute the unlevered cost of equity (r

_{U}) using β

_{U}= 0.79 in conjunction with r

_{F}= 3% (seen earlier in Section 4), and the market rate (r

_{M}) of 7.75% which is rate congruent with long-run historical market returns. Given r

_{U}, unlevered tax rate on equity (T

_{E1}) from Section 2.3.2, PBR (determined by trial and error to get 3.12%), the cash flow before taxes (CF

_{BT}) of $1,000,000, and the computation shown for g

_{U}(determined by PBR), we can calculate E

_{U}for nongrowth and growth with 3.12%. E

_{U}is important since each P choice refers to a debt choice defined as the proportion of E

_{U}retired with debt (D) or P = D/E

_{U}.

Alpha Computations for PTs: |

For the below PT computation, the unlevered tax rate on equity is T_{E1} and the levered tax rate on equity is T_{E2} where T_{E1} > T_{E2}. While T_{D} increases with leverage, it is only a levered tax rate as there is no debt when a firm is unlevered. Given the tax rates of T_{E}_{1} = 0.33, T_{E}_{2} = 0.283382, and T_{D} = 0.208669 at OCR, we have:α _{1} = (1 − T_{E}_{2})/(1 − T_{D}) = (1 − 0.283382)/(1 − 0.208669) = 0.905586α _{2} = (1 − T_{E}_{2})/(1 − T_{E}_{1}) = (1 − 0.283382)/(1 − 0.33) = 1.069579 |

Unlevered Firm Value Computations for PTs: |

Nongrowth (when g_{L} = 0% at OCR of Moody’s A3) where C is the before-tax payout to equity owners and the CSM uses a before-tax plowback ratio (PBR):PBR = 0 for nongrowth; C = (1 − PBR)(CF _{BT}) = (1 − 0)($1,000,000) = $1,000,000r _{U} = r_{F} + β_{U}(r_{M} − r_{F}) = 3% + 0.79(7.75% − 3%) = 6.7525%E _{U} (or V_{U}) = (1 − T_{E}_{1})C/r_{U} = (1 − 0.33)$1,000,000/0.067525 = $9,922,251 |

Growth (when g_{L} = 3.12% at OCR of Moody’s A3):PBR = 0.3435; retained earnings (RE) = PBR(CF _{BT}) = 0.3435($1,000,000) = $343,500; C = (1 − PBR)(CF_{BT}) = (1 − 0.3435)($1,000,000) = $656,500g _{U} = r_{U}(1 − T_{E}_{1})RE/C = 0.067525(1 − 0.33)$343,500/$656,500 = 2.3671807%; r_{Ug} = r_{U}–g_{U} = 6.7525% − 2.3671807% = 4.3853193%E _{U} (or V_{U}) = (1 − PBR)(1 − T_{E}_{1})CF_{BT}/r_{Ug} = (1 − 0.3435)(1 − 0.33)$1,000,000/0.043853193 = $10,030,170 |

## Appendix B. Pass-Through Application with under TCJA with Nongrowth

_{U}) retired with debt (D). The procedure to get costs of borrowing is described in Section 4. The nongrowth CSM given by Hull (2019) for PTs is used to compute the gain to leverage (G

_{L}). Firm value (V

_{L}) is unlevered firm value (E

_{U}) plus G

_{L}. Levered equity (E

_{L}) is V

_{L}minus debt (D). %∆E

_{U}is the percentage change in E

_{U}(or G

_{L}as a percent of E

_{U}). The net benefit from leverage (NB) is G

_{L}as a percent of D. The debt-to-firm value ratio (DV) is D/V

_{L}. The DV that is optimal is ODV and it is identified from the maximum G

_{L}(max G

_{L}) that coincides with the maximum V

_{L}(max V

_{L}) since max V

_{L}= E

_{U}+ max G

_{L}. As seen in the bold print (or optimal) column, the nongrowth optimal credit rating (OCR) is A3. While not shown because the last column in the table accompanying this appendix ends with a rating of B2, the violation of the nongrowth constraint does not occur until the Caa rating is reached. Where applicable, values are in millions of dollars and “n.a.” is not applicable.

Table to Accompany Appendix B: Pass-through Outputs with Nongrowth. | |||||||||||

P Choice = Proportion of Unlevered Firm Value (E_{U}) Retired by Debt (D) | |||||||||||

Variables | 0.0000 | 0.0641 | 0.1352 | 0.1674 | 0.2067 | 0.2582 | 0.2960 | 0.3069 | 0.3287 | 0.3227 | 0.3563 |

Credit rating | n.a. | Aaa | Aa2 | A1 | A2 | A3 | Baa2 | Ba1 | Ba2 | B1 | B2 |

D = P(E_{U}) | 0.000 | 0.636 | 1.342 | 1.661 | 2.051 | 2.562 | 2.937 | 3.045 | 3.261 | 3.202 | 3.535 |

Cost of debt: r_{D} | 3.00% | 3.63% | 3.78% | 3.98% | 4.08% | 4.22% | 4.56% | 5.00% | 5.40% | 6.51% | 7.21% |

Cost of equity: r_{L} | 6.75% | 7.08% | 7.23% | 7.43% | 7.53% | 7.67% | 8.01% | 8.45% | 8.85% | 9.96% | 10.7% |

Gain to leverage: G_{L} | 0.000 | 0.045 | 0.358 | 0.377 | 0.548 | 0.710 | 0.523 | 0.105 | −1.195 | −1.248 | −1.677 |

Firm value: V_{L} | 9.922 | 9.967 | 10.281 | 10.299 | 10.470 | 10.633 | 10.446 | 10.027 | 9.727 | 8.674 | 8.245 |

Equity value: E_{L} | 9.922 | 9.331 | 8.939 | 8.638 | 8.419 | 8.071 | 7.508 | 6.982 | 6.466 | 5.472 | 4.710 |

%∆E_{U} | 0.00% | 0.45% | 3.61% | 3.80% | 5.52% | 7.16% | 5.27% | 1.06% | −1.97% | −12.6% | −16.9% |

NB | 0.00% | 7.01% | 26.70% | 22.70% | 26.70% | 27.74% | 17.81% | 3.44% | −6.0% | −39.0% | −47.4% |

DV | 0.0000 | 0.0638 | 0.1305 | 0.1613 | 0.1959 | 0.2409 | 0.2812 | 0.3037 | 0.3353 | 0.3691 | 0.4287 |

Below we compute seven optimal outputs that are given in the bold print column where P is 0.2582 and the nongrowth OCR is A3. They are: D, max G_{L}, max V_{L}, E_{L}, max %∆E_{U}, NB, and ODV. For these computations, we use values up to ten decimal places to avoid rounding off errors. For example, 0.2582 is 0.25816244 as seen below for the optimal P choice. | |||||||||||

For the optimal P choice of 0.25816244, D retires 0.25816244 of E_{U} and we use the values of I = $136,498.62, r_{D} = 0.042168, and r_{L} = 0.076668 from Section 4 and the values of T_{D} = 0.20866933, α_{1} = 0.905585745, α_{2} = 1.0695787635, r_{U} = 0.067525, and E_{U} (or V_{U}) = $9,922,251.02 from Appendix A to get:D = P(E _{U}) = 0.25816244($9,922,251) = $2,561,553 or D = (1 − T_{D})I/r_{D} = (1 − 0.2086693)$136,498.62/0.042168 = $2,561,553Max G _{L} = [1 − α_{1}r_{D}/r_{L}]D – [1 − α_{2}r_{U}/r_{L}]E_{U} = [1 − 0.905585745(0.042168)/0.076668]$2,561,553 – [1 − 1.0695787635(0.067525)/0.076668]$9,922,251 = $1,285,696 – $575,225 = $710,471Max V _{L} = E_{U} + Max G_{L} = $9,922,251 + $710,471 = $10,632,722; E_{L} = V_{L} – D = $10,632,722.1 – $2,561,552.5 = $8,071,170Max %∆E _{U} = Max G_{L}/E_{U} = $710,471/$9,922,251 = 0.07160 or 7.16%NB = Max G _{L}/D = $710,471/$2,561,553 = 0.2774 or 27.74%; ODV = D/Max V_{L} = $2,561,553/$10,632,722 = 0.2409 |

## Appendix C. Pass-Through Application with under TCJA with Growth of 3.12%

_{U}) retired with debt (D). When using the CSM, growth is captured by the levered growth rate (g

_{L}) that is given as 3.12% in the optimal column in bold print. A growth rate of 3.12% is consistent with the historical growth in GDP for a recent seventy-year period as supplied by the US Bureau of Economic Analysis (2020). Given the definitions of g

_{L}for PTs and CCs as found in Hull (2019), the computation of g

_{L}requires an iterative process due to its interdependence with the gain to leverage (G

_{L}) as first noted by Hull (2010). The growth adjusted cost of levered equity (r

_{Lg}) is the levered equity rate of return (r

_{L}) minus g

_{L}. The growth CSM for PTs given by Hull (2019) is used to compute G

_{L}. Firm value (V

_{L}) is unlevered firm value (E

_{U}) plus G

_{L}. Levered equity (E

_{L}) is V

_{L}minus debt (D). %∆E

_{U}is the percentage change in E

_{U}(or G

_{L}as a percent of E

_{U}). The net benefit from leverage (NB) is G

_{L}as a percent of D. The debt-to-firm value ratio (DV) is D/V

_{L}. The DV that is optimal is ODV and it is identified from the maximum G

_{L}(max G

_{L}) that coincides with the maximum V

_{L}(max V

_{L}) since max V

_{L}= E

_{U}+ max G

_{L}. As seen in the bold print (or optimal) column of the table that accompanies this appendix, the growth optimal credit rating (OCR) is A3. The last two columns are shaded in gray to indicate that the growth constraint is violated. The violation of this constraint first occurs when a Moody’s credit rating of B1 is reached. This rating is a highly speculative rating. Violations tend to occur for high debt levels so that G

_{L}<0 and the firm can no longer fulfill its obligations to debt owners. Where applicable, values are in millions of dollars and “n.a.” is not applicable.

Table to Accompany Appendix C: Pass-through Outputs when Growth is 3.12%. | |||||||||||

heading | P Choice = Proportion of Unlevered Firm Value (E_{U}) Retired by Debt (D) | ||||||||||

Variables | 0.0000 | 0.0634 | 0.1338 | 0.1656 | 0.2045 | 0.2554 | 0.2929 | 0.3061 | 0.3279 | 0.3221 | 0.3557 |

Credit rating | n.a. | Aaa | Aa2 | A1 | A2 | A3 | Baa2 | Ba1 | Ba2 | B1 | B2 |

D = P(E_{U}) | 0.000 | 0.636 | 1.342 | 1.661 | 2.051 | 2.562 | 2.937 | 3.071 | 3.289 | 3.230 | 3.568 |

Levered growth rate: g_{L} | 2.37% | 2.57% | 2.69% | 2.85% | 2.96% | 3.12% | 3.47% | 3.92% | 4.41% | −4.95% | −4.70% |

Growth adjusted r_{L}: r_{Lg} | 4.39% | 4.51% | 4.54% | 4.58% | 4.57% | 4.55% | 4.54% | 4.53% | 4.44% | 14.91% | 15.36% |

Gain to leverage: G_{L} | 0.000 | 0.082 | 0.339 | 0.403 | 0.567 | 0.728 | 0.670 | 0.495 | 0.404 | −4.871 | −4.885 |

Firm value: V_{L} | 10.030 | 10.112 | 10.369 | 10.433 | 10.597 | 10.758 | 10.700 | 10.525 | 10.434 | 5.160 | 5.145 |

Equity value: E_{L} | 10.030 | 9.476 | 9.027 | 8.772 | 8.546 | 8.196 | 7.763 | 7.454 | 7.145 | 1.929 | 1.577 |

%∆E_{U} | 0.00% | 0.81% | 3.38% | 4.02% | 5.65% | 7.25% | 6.68% | 4.93% | 4.03% | −48.6% | −48.7% |

NB | 0.00% | 12.84% | 25.23% | 24.25% | 27.63% | 28.40% | 22.81% | 16.11% | 12.29% | −151% | −137% |

DV: D/V_{L} | 0.0000 | 0.0629 | 0.1294 | 0.1592 | 0.1936 | 0.2381 | 0.2745 | 0.2917 | 0.3152 | 0.6261 | 0.6934 |

Below we compute seven optimal outputs that are given in the bold print column where P is 0.2554 and the growth OCR is A3. They are: D, max G_{L}, max V_{L}, E_{L}, max %∆E_{U}, NB, and ODV. For these computations, we use values up to ten decimal places to avoid rounding off errors. For example, 0.2554 is 0.255384746 as seen below for the optimal P choice. | |||||||||||

For the optimal P choice of 0.255384746, D retires 0.255384746 of E_{U} and we use the values of I = $136,498.62, r_{D} = 0.042168, and r_{L} = 0.076668 from Section 4; g_{L} = 0.0312024037 (determined by iterative process using the gL equation for PTs given by Hull 2019); and, the values of T_{D} = 0.20866933, α_{1} = 0.9055857453, α_{2} = 1.0695787635, r_{U} = 0.067525, g_{U} = 0.0236718067, and E_{U} (or V_{U}) = $10,030,170 from Appendix A to get:D = P(E _{U}) = 0.255384746($10,030,170.38) = $2,561,553 or D = (1 − T_{D})I/r_{D} = (1 − 0.20866933)$136,498.623/0.042168 = $2,561,553.Using (2) with r _{Ug} = r_{U} – g_{U} = 0.067525 – 0.02367180674 = 0.04385319326; r_{Lg} = r_{L} – g_{L} = 0.076668 – 0.0312024037 = 0.0454655963, and above values for r_{D}, D, E_{U}, and r_{Ug}, we have:Max G _{L} = [1 − α_{1}r_{D}/r_{Lg}]D – [1 − α_{2}r_{Ug}/r_{Lg}]E_{U} = [1 − 0.9055857453(0.042168)/0.0454655963]$2,561,553 – [1 − 1.0695787635(0.04385319326)/0.0454655963]$10,030,170 = $410,094.2 – -$317,424.4 = $727,519Max V _{L} = E_{U} + Max G_{L} = $10,030,170 + $727,519 = $10,757,689; E_{L} = V_{L} – D = $10,757,689 – $2,561,553 = $8,196,136Max %∆E _{U} = Max G_{L}/E_{U} = $727,519/$10,030,170 = 0.07253 or about 7.25%NB = G _{L}/D = $727,519/$2,561,553 = 0.28401 or about 28.40%; ODV = D/Max V_{L} = $2,561,553/$10,757,689 = 0.2381 |

## References

- Asquith, Paul, and David Mullins Jr. 1986. Equity issues and offering dilution. Journal of Financial Economics 15: 61–89. [Google Scholar] [CrossRef]
- Baxter, Nevins. 1967. Leverage risk of ruin and the cost of capital. Journal of Finance 22: 395–403. [Google Scholar]
- Berk, Jonathan, Richard Stanton, and Josef Zechner. 2010. Human capital, bankruptcy and capital structure. Journal of Finance 65: 891–926. [Google Scholar] [CrossRef] [Green Version]
- Bowman, Woods. 2015. The Price of NPO Debt. NPO Quarterly. Available online: https://NPOquarterly.org/2015/08/06/the-price-of-NPO-debt/ (accessed on 31 December 2019).
- D’Melloa, Ranjan, and Mercedes Miranda. 2010. Long-term debt and overinvestment agency problem. Journal of Banking and Finance 34: 324–35. [Google Scholar] [CrossRef]
- Damodaran, Aswath. 2014. External and Internal Financing at US Firms, Page 30. Available online: http://pages.stern.nyu.edu/~adamodar/pdfiles/cfovhds/cfpacket2.pdf (accessed on 29 November 2019).
- Damodaran, Aswath. 2019. Damodaran Online: Home Page for Aswath Damodaran. Available online: http://pages.stern.nyu.edu/~adamodar/New_Home_Page/dataarchived.html (accessed on 25 November 2019).
- Damodaran, Aswath. 2020. Damodaran Online: Home Page for Aswath Damodaran. Available online: http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datacurrent.html (accessed on 15 January 2020).
- Dantzler, William. 2017. The Distinction between Partnership Debt and Partnership Equity. Tax Notes. Available online: https://www.whitecase.com/publications/article/distinction-between-partnership-debt-and-partnership-equity (accessed on 9 September 2019).
- DeAngelo, Harry, and Ronald Masulis. 1980. Optimal capital structure under corporate and personal taxes. Journal of Financial Economics 8: 3–29. [Google Scholar] [CrossRef]
- Donaldson, Gordon. 1961. Corporate Debt Capacity: A Study of Corporate Debt Policy and the Determination of Corporate Debt Capacity. Boston: Harvard Business School, pp. 1–294. [Google Scholar]
- Farrar, Donald, and Lee Selwyn. 1967. Taxes, corporate financial policy, and returns to investors. National Tax Journal 20: 444–54. [Google Scholar]
- Federal Reserve Economic Data. 2019. Moody’s Seasoned Baa Corporate Bond Yield. Federal Reserve Bank of St. Louis. Available online: https://fred.stlouisfed.org/series/BAA (accessed on 15 January 2019).
- Florida, Richard, and Donald Smith. 1993. Keep the government out of venture capital. Issues in Science and Technology 9: 61–68. [Google Scholar]
- Gay, Gerald, and Jouahn Nam. 1998. The underinvestment problem and corporate derivatives use. Financial Management 27: 53–69. [Google Scholar] [CrossRef] [Green Version]
- Graham, John. 2000. How big are the tax benefits of debt? Journal of Finance 55: 1901–41. [Google Scholar] [CrossRef]
- Graham, John, and Campbell Harvey. 2001. The theory and practice of corporate finance: Evidence from the field. Journal of Financial Economics 60: 187–243. [Google Scholar] [CrossRef]
- Graham, John, and Mark Leary. 2011. A review of empirical capital structure research and directions for the future. Annual Review of Financial Economics 3: 309–45. [Google Scholar] [CrossRef]
- Grant Thorton International Ltd. 2019. Should Pass-through Owners Switch to C Corp Status? Available online: https://www.grantthornton.com/library/articles/tax/2018/switch-C-corp-status.aspx (accessed on 11 March 2020).
- Henrekson, Magnus, and Tino Sanandaji. 2011. Entrepreneurship and the theory of taxation. Small Business Economics 37: 167–85. [Google Scholar] [CrossRef]
- Hodge, Scott. 2014. The US Has More Individually Owned Businesses than Corporations. Available online: https://taxfoundation.org/us-has-more-individually-owned-businesses-corporations (accessed on 21 January 2020).
- Houston, Dana. 2019. You’ve Decided to Switch to a C Corporation: Here’s What’s Next. Available online: https://www.claconnect.com/resources/articles/2019/youve-decided-to-switch-to-a-c-corporation-heres-whats-next (accessed on 12 May 2010).
- Hull, Robert. 2010. A capital structure model with growth. Investment Management and Financial Innovation 7: 26–40. [Google Scholar]
- Hull, Robert. 2012. A capital structure model with wealth transfers. Investment Management and Financial Innovation 9: 19–32. [Google Scholar]
- Hull, Robert. 2014. A Capital Structure Model (CSM) with tax rate changes. Investment Management and Financial Innovations 11: 8–21. [Google Scholar]
- Hull, Robert. 2018. Capital Structure Model (CSM): Correction, constraints, and applications. Investment Management and Financial Innovation 15: 245–62. [Google Scholar] [CrossRef] [Green Version]
- Hull, Robert. 2019. Business wealth and tax policy. Theoretical Economics Letters 19: 1020–39. [Google Scholar] [CrossRef] [Green Version]
- Hull, Robert. 2020. Credit ratings and firm value. Investment Management and Financial Innovation 17: 157–68. [Google Scholar] [CrossRef]
- Hull, Robert, and David Price. 2015. Pass-through valuation. The Journal of Entrepreneurial Finance 17: 82–116. [Google Scholar]
- Hull, Robert, Sungkyu Kwak, and Walker Rosemary. 2018. Hedge fund variables and short-run SEO returns. International Journal of Managerial Finance 14: 322–41. [Google Scholar] [CrossRef]
- Jensen, Michael. 1986. Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76: 323–29. [Google Scholar]
- Jensen, Michael, and William Meckling. 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3: 305–60. [Google Scholar] [CrossRef]
- Kisgen, Darren. 2006. Credit ratings and capital structure. Journal of Finance 61: 1035–72. [Google Scholar] [CrossRef]
- Kisgen, Darren. 2009. Do firms target credit ratings or leverage levels? Journal of Financial and Quantitative Analysis 44: 1323–44. [Google Scholar] [CrossRef] [Green Version]
- Korteweg, Arthur. 2010. The net benefits of leverage. Journal of Finance 65: 2137–70. [Google Scholar] [CrossRef]
- Kraus, Alan, and Robert Litzenberger. 1973. A State-preference model of optimal financial leverage. Journal of Finance 28: 911–22. [Google Scholar] [CrossRef]
- Leland, Hayne. 1998. Agency costs, risk management, and capital structure. Journal of Finance 53: 1213–43. [Google Scholar] [CrossRef]
- Masulis, Ronald. 1983. The impact of capital structure change on firm value: Some estimates. Journal of Finance 38: 107–26. [Google Scholar] [CrossRef]
- Miller, Merton. 1977. Debt and taxes. Journal of Finance 32: 261–75. [Google Scholar]
- Modigliani, Franco, and Merton Miller. 1963. Corporate income taxes and the cost of capital: A correction. American Economic Review 53: 433–43. [Google Scholar]
- Mohaghegh, Mohsen. 2019. Evolution of Wealth Inequality In the US: Entrepreneurship and Financial Intermediation, Working Paper, Columbus, Ohio, Department of Economics, Ohio State University. Available online: https://drive.google.com/file/d/1lnkax3Vz6Ln4awmQsJNTMotmFxgiP_kk/view (accessed on 26 February 2020).
- Morningstar. 2019. Credit Ratings. Available online: https://ratingagency.morningstar.com/mcr/ratings-surveillance/corporate-financial%20institutions (accessed on 21 July 2019).
- Myers, Stewart. 1977. Determinants of corporate borrowing. Journal of Financial Economics 5: 147–75. [Google Scholar] [CrossRef] [Green Version]
- Myers, Stewart, and Nicholas Majluf. 1984. Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics 13: 187–221. [Google Scholar] [CrossRef] [Green Version]
- National Federation of Independent Business. 2013. Entity Choice and Effective Tax Rates. Available online: http://www.s-corp.org/wp-content/uploads/2013/08/Quantria_Study_ETR_8-6-13_Final_pm.pdf (accessed on 7 October 2019).
- Penn Wharton Budget Model. 2017. The Tax Cuts and Jobs Act, as Reported by Conference Committee, December 15. Available online: https://budgetmodel.wharton.upenn.edu/issues/2017/12/15/effective-tax-rates-by-industry (accessed on 26 October 2019).
- Robert, Celene. 2019. Should You Convert Your Company to a C Corp? Available online: https://www.indinero.com/blog/converting-your-company-to-a-c-corp (accessed on 28 April 2020).
- SBA. Office of Advocacy. 2009. Effective Federal Income Tax Rates Faced By Small Businesses in the United States. Small Business Administration. April. Available online: https://www.sba.gov/sites/default/files/advocacy/rs343tot.pdf (accessed on 28 October 2019).
- Tax Foundation. 2014. The U.S. Corporate Effective Tax Rate: Myth and the Fact. Available online: https://files.taxfoundation.org/legacy/docs/SR214.pdf (accessed on 28 October 2018).
- Tax Foundation. 2019. Pass-Through Businesses Q&A. Available online: https://taxfoundation.org/pass-through-businesses/ (accessed on 9 January 2020).
- Tax Policy Center. 2016. Table T16-0181: Distribution of Tax Units with Business Income by Expanded Cash Income Percentile, 2016. August 23. Available online: https://www.taxpolicycenter.org/model-estimates/distribution-business-income-august-2016/t16-0182-distribution-business-income (accessed on 18 January 2020).
- Tax Policy Center. 2018. How Might the Tax Cuts and Jobs Act Affect Economic Output? Available online: https://www.taxpolicycenter.org/briefing-book/how-might-tax-cuts-and-jobs-act-affect-economic-output (accessed on 18 October 2018).
- US Bureau of Economic Analysis. 2020. US Real GDP Growth Rate by Year. Available online: http://www.multpl.com/us-real-gdp-growth-rate/table/by-year (accessed on 11 January 2020).
- Van Binsbergen, Jules, John Graham, and Jie Yang. 2010. Cost of debt. Journal of Finance 65: 2089–136. [Google Scholar] [CrossRef]

1 | By before-tax, we mean after all expenses (including replacement costs) have been paid except for federal tax expenses. Thus, expenses include all applicable non-federal taxes (such as state taxes, payroll taxes, property taxes, sales taxes, and so forth). Since we begin with an unlevered firm, interest on debt is not yet an expense. |

2 | According to our records, Damodaran’s ICRs were the same for 2018 and 2019. Thus, whereas spreads changed for these two years, the ratings and ICRs that are matched to the spreads did not change. We do not know to what extent, if any, ICRs were provided prior to 2018 because Damodaran’s archive files do not always contain the same details that were provided during the year when the data was first reported. Regardless, our best guest is that ICRs for 2017 existed and are similar to 2018 and 2019. Prior to 2017, we have no record that Damodaran reported ICRs. |

**Figure 1.**Nongrowth Pass-Through Gain to Leverage (G

_{L}) Versus debt-to-firm value ratios (DVs) under Tax Cuts and Job Acts (TCJA) with g

_{L}of 0% and optimal credit rating (OCR) of A3. Gain to Leverage (G

_{L}), solid line; 1st Component, dotted line, and 2nd Component, dashed line.

**Figure 2.**Growth Pass-Through G

_{L}Versus DV under TCJA with g

_{L}of 3.12% and OCR of A3. Gain to Leverage (G

_{L}), solid line; 1st Component, dotted line, and 2nd Component, dashed line.

**Figure 3.**Nongrowth firm value (V

_{L}) (dotted line), 3.12% Growth V

_{L}(smaller dashed line), 3.90% Growth V

_{L}(solid line), and 4.50% Growth V

_{L}(larger dashed line) Versus Moody’s Credit Ratings for Pass-Throughs under TCJA with OCR of A3 target except for top trajectory where OCR is Baa2.

P Choice | ICR | Interest (I) | DV | Spread | Ratings | r_{D} | r_{L} |
---|---|---|---|---|---|---|---|

0.0641 | 24.000 | $28,329 | 0.0638 | 0.630% | Aaa | 3.63% | 7.08% |

0.1352 | 11.000 | $62,682 | 0.1305 | 0.780% | Aa2 | 3.78% | 7.23% |

0.1674 | 8.500 | $82,214 | 0.1613 | 0.975% | A1 | 3.98% | 7.43% |

0.2067 | 6.750 | $104,867 | 0.1959 | 1.076% | A2 | 4.08% | 7.53% |

0.2582 | 5.250 | $136,499 | 0.2409 | 1.217% | A3 | 4.22% | 7.67% |

0.2960 | 4.250 | $170,616 | 0.2812 | 1.560% | Baa2 | 4.56% | 8.01% |

0.3069 | 3.750 | $195,564 | 0.3037 | 2.000% | Ba1 | 5.00% | 8.45% |

0.3287 | 3.250 | $228,112 | 0.3353 | 2.400% | Ba2 | 5.40% | 8.85% |

0.3227 | 2.750 | $272,409 | 0.3691 | 3.510% | B1 | 6.51% | 9.96% |

0.3563 | 2.250 | $336,289 | 0.4287 | 4.212% | B2 | 7.21% | 10.66% |

0.4054 | 1.750 | $436,543 | 0.5201 | 5.148% | B3 | 8.15% | 11.60% |

0.3751 | 1.375 | $560,751 | 0.6168 | 8.200% | Caa | 11.20% | 14.65% |

0.4833 | 1.025 | $758,928 | 0.7942 | 8.642% | Ca2 | 11.64% | 15.09% |

0.6173 | 0.650 | $1,207,022 | 1.1752 | 11.341% | C2 | 14.34% | 17.79% |

0.8333 | 0.380 | $2,081,650 | 1.9225 | 15.116% | D2 | 18.12% | 21.57% |

_{U}) retired with debt (D). Damodaran’s data matches fifteen credit spreads and ratings to fifteen ranges of interest coverage ratios (ICRs), where the latter are given for three firm classifications of small, large, and financial service. For this study, we used the ICRs from the small firm and large firm classifications, where large refers to firms with assets over five billion dollars and small refers to those with assets under five billion dollars. Since PTs are characterized by small US firms, we applied the small ICRs to PTs. Similarly, we applied the large ICRs to CCs since they contain the majority of the largest US firms. For the below computations, we used the average of each ICR range except for the first and last ranges that provide extremely large positive or negative ranges and would give average values that are impractical. For these two ranges, we had to assign realistic values that still fall within these ranges in a manner described by Hull (2020). Although a more common definition of ICR is EBIT/I, Damodaran uses the definition of ICR = (1 − T)EBIT/I where T is the average tax rate on business income, EBIT is earning before interest and taxes, and I is interest. For PTs, T is the effective levered business tax rate paid at the personal tax level. Thus, T is the same as T

_{E}

_{2}used in the PT equations given by Hull (2019) and illustrated in Appendix A. For CCs, T is the corporate tax rate used in the CC equations given by Hull (2018). In terms of the CSM, EBIT is analogous to CF

_{BT}. To compute I per $1,000,000 in perpetual CF

_{BT}for PTs, we rearranged ICR = (1 − T)EBIT/I by inserting T

_{E}

_{2}for T and CF

_{BT}for EBIT to get I = (1 − T

_{E2})CF

_{BT}/ICR. Using the I values, we calculated corresponding D values using D = (1 − T

_{D})I/r

_{D}where T

_{D}is the effective tax rate on debt (D) described in Section 2.3 and r

_{D}is the cost of debt as computed in this table. Given D, we computed a P choice as P = D/E

_{U}where E

_{U}is computed in Appendix A and DV is D/V

_{L}where V

_{L}is firm value and DV is first computed in Appendix B. The bold print row contains optimal values including the optimal DV (ODV) and optimal credit rating (OCR). The last four gray-shaded rows represent unfeasible values, as the CSM nongrowth constraint is violated. To get the costs of debt (r

_{D}), we used r

_{D}= spread + r

_{F}where r

_{F}is the risk-free rate of 3.00%, which is a value consistent with long-term government bonds the past ten years. We computed the borrowing cost of levered equity (r

_{L}) as r

_{L}= r

_{D}+ EPB where EPB refers to an equity premium over a bond return. Damodaran (2020) and the Federal Reserve Economic Data (2019) indicate an EPB of 3.45%. While this paper’s EPB uses data for 2019 as given by Damodaran (2020), it is also consistent with data for 2018 as supplied by Damodaran (2019).

P Choice | ODV | E_{U} | Max V_{L} | Max G_{L} | Max %ΔE_{U} | NB | |
---|---|---|---|---|---|---|---|

Panel A. Pass-Throughs (PTs) | |||||||

Nongrowth: Pre-TCJA | 0.2559 | 0.2375 | $9.626 | $10.371 | $0.745 | 7.74% | 30.24% |

Growth (g_{L} = 3.12%): Pre-TCJA | 0.2556 | 0.2371 | $9.638 | $10.390 | $0.752 | 7.80% | 30.51% |

Nongrowth: TCJA | 0.2582 | 0.2409 | $9.922 | $10.633 | $0.710 | 7.16% | 27.74% |

Growth (g_{L} = 3.12%): TCJA | 0.2554 | 0.2381 | $10.030 | $10.758 | $0.728 | 7.25% | 28.40% |

Growth (g_{L} = 3.90%): TCJA | 0.2407 | 0.2226 | $10.644 | $11.506 | $0.863 | 8.11% | 33.68% |

Growth (g_{L} = 4.50%): TCJA | 0.2718 | 0.2499 | $10.807 | $11.756 | $0.948 | 8.77% | 32.28% |

Overall Average for PTs | 0.2563 | 0.2377 | $10.111 | $10.902 | $0.791 | 7.81% | 30.48% |

Panel B: C Corps (CCs) | |||||||

Nongrowth: Pre-TCJA | 0.5434 | 0.4354 | $8.038 | $10.032 | $1.994 | 24.81% | 45.65% |

Growth (g_{L} = 3.12%): Pre-TCJA | 0.5501 | 0.4609 | $7.940 | $9.476 | $1.536 | 19.35% | 35.17% |

Nongrowth: TCJA | 0.5288 | 0.4537 | $9.769 | $11.385 | $1.616 | 16.54% | 31.28% |

Growth (g_{L} = 3.12%): TCJA | 0.5124 | 0.4526 | $10.080 | $11.413 | $1.333 | 13.22% | 25.81% |

Growth (g_{L} = 3.90%): TCJA | 0.4941 | 0.4337 | $10.454 | $11.910 | $1.455 | 13.92% | 28.18% |

Growth (g_{L} = 4.50%): TCJA | 0.4766 | 0.4132 | $10.839 | $12.502 | $1.663 | 15.35% | 32.20% |

Overall Average for CCs | 0.5176 | 0.4416 | $9.520 | $11.120 | $1.600 | 17.20% | 33.05% |

_{U}) retired with debt (D). As argued in Section 4.2, the optimal credit rating (OCR) is Moody’s A3 for PTs with an exception being for the 4.50% growth tests where OCR is Baa2. For CCs, OCR is Baa2 for all tests. This OCR is based on the same considerations given in Section 4.2 when identifying an OCR for PTs. All tests have a before-tax cash flow of $1,000,000. All dollar values in this Table 2 are given in millions. The last row of each panel provides averages of the preceding six rows. The averages for max %∆E

_{U}in Panel B are high compared to pre-TCJA research. These higher values can be explained by the fact that outputs can be sensitive to the change in credit spreads over time. Tests using Damodaran’s archived spreads for 2018 at http://pages.stern.nyu.edu/~adamodar/New_Home_Page/dataarchived.html provide lower max %∆E

_{U}values that are consistent with pre-TCJA research. For example, while the average of all max %∆E

_{U}values for both panels in Table 2 is 12.50%, it is 7.33% using 2018 data.

© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Hull, R.
Pass-Through and C Corp Outputs under TCJA. *Int. J. Financial Stud.* **2020**, *8*, 46.
https://doi.org/10.3390/ijfs8030046

**AMA Style**

Hull R.
Pass-Through and C Corp Outputs under TCJA. *International Journal of Financial Studies*. 2020; 8(3):46.
https://doi.org/10.3390/ijfs8030046

**Chicago/Turabian Style**

Hull, Robert.
2020. "Pass-Through and C Corp Outputs under TCJA" *International Journal of Financial Studies* 8, no. 3: 46.
https://doi.org/10.3390/ijfs8030046