Leveraged buyouts and tax incentives.

* Leveraged buyout activity increased from 99 transactions worth $3 billion in 1981 [18] to peak 338 transactions worth over $61 billion in 1989 [19]. Suggested motivating forces for corporate acquisitions include non-value-maximizing behavior on the part of managers, the free cash flow problem, tax incentives, deregulation, synergies, economies of scale and scope, and increasing globalization of U.S. markets ([2], [9], [10], [13], and [22]). However, of this list, tax incentives are the most frequently discussed motivation for corporate acquisitions, especially in leveraged buyouts ([4], [8], [9], [10], [13], [14], and [20]). Kaplan [14] estimates the median value of tax benefits in management buyouts of public companies to be anywhere from 21% to 143% of the premium paid to pre-buyout shareholders. Hayne [8] shows the use of tax shields, that would not be fully utilized in the absence of an acquisition, is significant in explaining the gains to target firms' shareholders as well as to the acquiring firms. The step-up in the basis of asset, in particular, is significant in explaining these gains in taxable acquisitions (Hayne [18]).

The purpose of this study is to investigate the tax incentives to managers of leveraged buyouts by estimating the change in taxes paid by a firm due to a leveraged buyout. Kaplan [14] estimated the tax change for management buyouts according to tax rules existing before the 1986 Tax Reform Act (TRA '86) and on the basis of static rules, such as the repayment of debt in eight equal principal payments. Our study estimates that change in taxes paid by 23 of the largest leveraged buyouts in the years 1988, 1989 and 1990. The estimates are made under four different tax structures. First, the change is estimated according to two possible acquisition structures allowed under TRA '86 rules. Then, it is estimated again assuming the buyout can be structured to preserve the "General Utilities" treatment repealed by TRA '86. Finally, as a reference point, the change in taxes is estimated under a structure that would mimic pre-TRA '86 rules. Under each structure, the reduction in taxes to the leveraged buyout is expressed as a percentage of the acquisition premium and also as a present value at the time of the buyout. The first section of the paper briefly explains tax treatment of leveraged buyouts and describes the significant changes brought about by the TRA '86. The second section explains the method used to estimate the reduction in taxes paid by a firm due to a leveraged buyout. The third section presents the results of the estimation, and the fourth section offers a summary and conclusions.

I. Tax Background

Before TRA '86, the General Utilities [7] doctrine allowed no gain to be recognized by a corporation when it distributed assets in a liquidation under Section 336 of the Internal Revenue Code. Further, by suitable restructuring of the acquisition under Section 338, and despite nonrecognition of a gain, the acquiring corporation could step up the basis of the acquired assets, thus securing increased depreciation and reducing future taxes.

With congressional belief that the taxpayer was subsidizing leveraged buyouts, TRA ;86 repealed the General Utilities doctrine. Section 336 now requires that a gain be recognized by a liquidating corporation and in order to obtain a step-up in basis, achieved by a Section 338 election, there is a payment of capital gains taxes levied at the target level. Because TRA '86 also repealed the corporate capital gains tax, the applicable tax rate rose from 28% on an actual sale to 34% on a deemed or actual sale. After TRA '86,(1) in the simplest example with a stock sale and a Section 338 election, the tax payable would be 62% on the increment in value (34% on corporate gain plus 28% on stockholders as they are cashed out by the leveraged buyout). With other straightforward structures, such as a stock sale with no Section 338 election, the tax would be 28% (payable by the stockholders -- but the corporation could get no step-up in basis), and with an asset acquisition followed by complete liquidation (under Section 337), the tax would be 52.48% (34% at the corporate level plus 28% on the net proceeds at the stockholder level).

Another adverse change to leveraged buyout planners in TRA '86 was the new procedure for allocating the purchase price to the basis of acquired assets. Previously, a so-called second tier allocation process was allowed, which could result in individual depreciable assets being stepped up to more than their fair market value, thereby reducing the amount of the total purchase price allocated to undepreciable goodwill. New regulations written pursuant to TRA '86 require allocation into four classes of assets by prescribed priority, where goodwill is the fourth priority -- thus, goodwill received the whole of the modified aggregate deemed sale price less only the fair market value of assets in Classes I through III. This study allocates the gain using a Treasury-acceptable procedure following that laid out in Bittker and Eustice [3] and Bush and Mullany [6], with the added simplification (since we do not know the fair market value of the individual asset classes) that the book values of Classes I and II (primarily cash and marketable securities) are equal to their fair market values, leaving the unallocated part of the adjusted sales price to be distributed proportionately to book values for assets in Classes III and IV. Scholes and Wolfson [21], in their appendix, more completely explain tax law changes for mergers and acquisitions resulting from TRA '86.

II. Sample and Method

This study estimates taxes paid by 23 of the largest corporate leveraged buyouts in the years 1988, 1989 and 1990. Exhibit 1 describes the sample.(2) Our sample was obtained by beginning with a list of the largest 100 acquisitions for each of the three years. Sixty-two of the 300 were leveraged buyouts and 39 of these were excluded because the targets were divisions rather than listed corporations or due to other lack of data.

The change in taxes paid by a firm because of a leveraged buyout is calculated under four possible tax structures:

(i) TRA '86 with no step-up in the basis of assets (no Section 338 election). Thus, there is no increased depreciation tax shield, but also no capital gains tax is levied on the target firm. There may be a small ordinary gain due to depreciation recapture. The tax incentive is the additional interest tax shield.

(ii) TRA '86 with a step-up in the basis of assets (Section 338 election). Under this tax structure, there is an increase in the depreciation tax shield as well as the interest tax shield, but there will be a large ordinary gain and a capital gain assessed on the target firm.

(iii) TRA '86 with a Section 338 election and the preservation of "General Utilities" treatment. Under this tax structure, there is an increase in the depreciation and interest tax shields, but it assumes the acquisition can be structured in such a way as to avoid the capital gains tax levied on the target firm. As with structure (i) above, there may be a small ordinary gain.

(iv) Pre-TRA '86 with a Section 338 election (no capital gains tax). Under this tax structure, there is an increase in the depreciation and interest tax shields. There is no capital gains tax levied on the target firm. Due to the higher tax rate, the ordinary gain will also be taxed at the 46% rather than 34% rate.

It is reported that firms will not choose the second structures because of the substantial capital gains taxes paid under this option (e.g., Kaplan [14] and Lynch, Baldasaro and Siegel [17]). This tax structure is examined in order to quantify this belief. The third structure is considered because there are persistent hints in the literature ([15], [15], [16], and [17]) that by careful restructuring, the resurrection of "General Utilities" is possible. The fourth structure is considered as a base case and to allow comparison with previous studies. The results will answer three important questions. First, what are the tax incentives encouraging leveraged buyouts? Second, what impact did TRA '86 have upon the tax incentives of leveraged buyouts? Third, is the option to step up the basis of assets under TRA '86 really inferior to not stepping up the basis?

Reduction in taxes paid due to a leveraged buyout is estimated by forecasting future taxable income and calculating taxes paid at the firm's actual marginal corporate income tax rate. This estimate of future taxes paid is calculated under four tax structures for each of the 23 firms in the sample. First, taxes paid are estimated assuming the leveraged buyout occurs under one of the four tax structures described above. Then, taxes paid are estimated assuming the firm continues without a leveraged buyout. The difference between taxes paid with the leveraged buyout and the taxes paid without a leveraged buyout is the measure of the change in taxes paid due to a leveraged buyout under each of the tax structures.

The process begins by estimating the tax on the recapture of accelerated depreciation as well as a capital gains tax on the target firm, where appropriate, in each of the four tax structures. The estimates of future taxes paid begin with a forecast of earnings before taxes (EBT) for each firm. Each firm's EBT for the last fiscal year prior to the buyout is forecast forward using Value Line Investment Survey's estimate of future earnings growth for that particular firm. The no-leveraged-buyout scenario grows EBT forward at the forecasted growth rate. The leveraged-buyout scenarios change the forecasted EBTs by the estimated increase in interest expense(3) and depreciation expense resulting from the leveraged buyouts. Each firm's taxes paid are calculated based upon the forecasted EBT and the actual corporate income tax rate based upon each firm's taxable income level. We assume that all cash flow is used to pay down debt from the leveraged buyout, consistent with the reported behavior of leveraged buyouts. Also, a target firm is assumed held for five or eight years before being sold.(4)

For each of the 23 leveraged buyouts, taxes paid under each of the four tax structures are forecast out for five and eight years and then discounted to the buyout time using each company's before-tax cost of debt.(5) Then, assuming no leveraged buyout took place, taxes paid on earnings are forecast for each of the 23 firms for five and eight years and similarly discounted to the earlier time period. For each of the four structures, a reduction in the present value of taxes paid due to a leveraged buyout can be calculated by taking the difference between the present value of taxes paid with a leveraged buyout and the present value of taxes paid without a leveraged buyout. The difference is calculated for a five-year horizon and an eight-year horizon.

In summary, the change in taxes is the net result of changes in the following:

(i) Taxes on corporate income, as impacted by the interest tax shield on buyout debt, and, where applicable, the increased depreciation from a step-up in the basis of assets.

(ii) Ordinary gains tax on the recapture of accelerated depreciation .(6)

(iii) Capital gains tax levied on the target firm, where applicable. It is important to note that this tax is paid by the firm, not the firm's shareholders. This study is concerned with managerial incentives; thus, taxes on exiting stockholders are ignored.

Besides updating Kaplan's [14] study with the TRA '86 law, this study advances Kaplan's work by the following:

(i) Debt repayment is based upon each firm's projected ability to pay. All earnings after taxes are used to repay new debt. Kaplan [14] estimates two scenarios. The first assumes new debt is never reduced. The second assumes new debt is repaid in eight equal annual principal payments regardless of ability to pay.

(ii) The interest on debt is the actual rate. Kaplan [14] uses a constant (12.5%) rate for all firms in the sample.

(iii) The actual tax benefits from stepping up the basis of assets are estimated for each firm. Kaplan [14] assumes the value of the tax shelter from stepping up the value of assets equals 26% of the stepped-up value.

(iv) The ability to use tax shields each year is forecast and excess shields are rolled forward until they can be used. Kaplan [14] assumes all deductions can be used in the year generated.

The use of the same earnings forecast for both the buyout scenario and the nonbuyout scenario may bias our estimates. If management efficiency is a motivating force in buyouts, then a buyout should lead to higher operating income and a greater tax liability. This will bias the estimates of the tax changes. Although we do not explicitly incorporate this increase management efficiency in our estimates, the increase in earnings growth necessary for the leveraged buyout to utilize all available tax shields is calculated and presented in Exhibit 4.

III. Estimates of Tax Reductions

Exhibit 2 shows the average estimated change in the present value of taxes paid over a five-year horizon by 23 leveraged buyouts. The change is relative to the estimated present value of taxes paid if the firms were not acquired. Each row of the exhibit shows the change in taxes under one of the four tax structures described in the previous section. Each change in the exhibit is presented twice. First, the change in tax payments is shown as a percentage of the acquisition premium. It is unweighted average of the 23 firms. The premium is calculated as the difference between the final purchase price of the equity and the market price two months prior to the initial announcement.(7) Second, the average change in present value dollars is shown. The first column of Exhibit 2 shows the average reduction in corporate income taxes paid due to added depreciation and interest tax shields created by the leveraged buyout. The second column shows the average additional taxes paid on a leveraged buyout from recapture of accelerated depreciation and a capital gains tax on the target firm, where appropriate. The third column shows the total leveraged buyout impact on taxes by combining the first two columns. The fourth column shows the average additional taxes paid by a leveraged buyout under a TRA '86 tax structure when compared to the pre-TRA '86 tax structure. Exhibit 3 is identical to Exhibit 2, except for the use of an eight-year horizon.

The first row of Exhibit 2 shows the pre-TRA '86 tax structure examined in the previously cited studies. Thus, operating taxes reduce, on average, by 57.4% of the acquisition premium, while taxes on gains amount to 9.0% of the premium, leaving a net incentive to the leveraged buyout managers equal to 48.4% of the premium. Our results are broadly consistent with those of earlier studies. The first column of Exhibit 2 highlights the impact of the reduction in the marginal corporate tax rate from 45% to 34% in 1986 -- the pre-TRA '86 scenario has at least an extra $112 million ($415.6 - $303.6) reduction in taxes in present value terms when compared to any TRA '86 tax structure. The small differences in the present value of taxes in the first column for the three TRA '86 tax structures are due to the inability of the leveraged buyouts to fully use the tax shields within the five-year horizon. The second column of Exhibit 2 shows the relatively minor impact of accelerated depreciation recapture and the huge impact of the taxes levied on the target firm if a Section 338 election is made (TRA '86 with step-up). The third column nets out the first two columns. The present value of tax reductions under the pre-TRA '86 structure equals 48.4% of the acquisition premium with a five-year horizon. Under TRA '86 tax structures with a step-up, it averaging 35.9% to 32.6% of the acquisition premium, and with a Section 338 election, taxes represent a positive cost equal to 27.8% of the premium. The fourth column shows the additional taxes paid under a TRA '86 tax structure relative to the pre-TRA '86 tax structure. When compared with the pre-TRA '86 scenario, the average leveraged buyout pays between $103.8 million and $613.2 million extra in taxes because of TRA '86.

Because the leveraged buyouts were typically unable to fully utilize the available tax shields within the five-year horizon, the estimates were repeated with an eight-year horizon. These results are in Exhibit 3. The second row shows the average change in taxes, assuming leveraged buyouts occur under the TRA '86 (no step-up) structure. Operating taxes reduce, on average, by 51.5% of the acquisition premium, while taxes on gains amount to 6.7% of the premium, leaving a net incentive to the leveraged buyout managers equal to 44.8% of the premium, which is less than the incentive under the pre-TRA '86 tax structure by a amount equal to 26.4% of the premium. There are now clear differences in taxes on operating income in TRA '86 tax structures, as shown in the first column. For example, the increased depreciation from stepping up the basis of assets reduces operating income taxes by a present value of $69.5 million in TRA '86 tax structures. Thus, TRA '86 (no step-up) has the smallest reduction in taxes on operating income because the leveraged buyout creates no additional depreciation tax shield under this tax structure. TRA '86 (with step-up) has a huge increase in taxes on the target company because the leveraged buyout causes a large capital gains tax to be levied under this tax structure. The third column of Exhibit 3 shows that the present value of tax reductions under the pre-TRA '86 structure equals 71.2% of the acquisition premium with an eight-year horizon. Under TRA '86 tax structures, it is now, at best, averaging 52.2% to 44.8% of the acquisition premium, and with a Section 338 election, taxes represent a positive cost equal to 11.5% of the premium. These differences now produce changes in net taxes that differ from the pre-TRA '86 scenario by between $164.6 million and $674 million as shown in the fourth column. When comparing Exhibit 3 to Exhibit 2, the additional operating income achieved in the extra three years with an eight-year horizon makes an appreciable difference in the ability to use tax shields.

Although not apparent in Exhibit 3, the calculations show that not all leveraged buyouts use their increased tax shields by the end of eight years, given the earnings as forecast. But, leveraged buyouts may lead to improved earnings due to better operations and the pressures brought on by increased financial leverage (Baker and Wruck [2] and Jensen, Kaplan and Stiglin [11]). The estimates up to this point have not allowed for any additional earnings growth due to the leveraged buyout.(8) We incorporate this possibility in Exhibit 4 by estimating the average earnings growth rate required by a buyout in order to use the increased tax shields within the eight-year horizon.

The first column of Exhibit 4 shows the average Value Line earnings growth forecast for the 23 firms. These are the earnings growth rates used to calculate the results shown in the first two exhibits. The last column has the unweighted average annual earnings growth rate required for the leveraged buyout to use all the available tax shields within the eight-year horizon. There are three rows under each of the three TRA '86 tax structures presented in the exhibit. The first row shows the average for all 23 firms in the sample. The second row shows the average for those firms using up all the increased tax shields within eight years without earnings growing at a faster rate than forecast by Value Line. The third row under each tax structure shows the average for those firms not using all the increased tax shields within eight years. In other words, under the no-step-up tax structure, 11 leveraged buyouts' earnings would not sufficiently large to use all their increased tax shields at Value Line's forecasted earnings growth rate. The average growth rate would need to be 28.5%, instead of the 11.2% forecast for the firms to use all available tax shields. For this subsample of firms not using all the increased tax, shields, earnings growth would have to more than double and almost triple the forecast growth rate in order to use all tax shields in eight years. While management may have the potential to cause substantial increases immediately after the leveraged buyout, it may be doubted whether growth rates on the order of 30% annually for eight years are achievable.

IV. Summary and Conclusions

The major studies of leveraged buyouts and the effect of taxes on management behavior were based upon pre-TRA '86 tax structures. TRA '86 created a major change in leveraged buyout tax structures as summarized in Section I of this paper. The revisions are sufficiently extensive as to require a thorough reexamination of leveraged buyout tax incentives. In addition to examining the change in tax structures, this study advances the literature by using a comprehensive simulation approach, rather than a more static estimation procedure. Results, based on 23 of the largest leveraged buyouts between 1988 and 1990, show that the congressional intent to reduce tax benefits was achieved. With an eight-year horizon, leveraged buyouts in the study would have suffered extra taxes averaging $164.6 million, $234.1 million or $674 million depending upon which TRA '86 tax structure was employed. The five-year horizon simulation showed less change ($103.8 million, $120.9 million and $613.2 million) since the five years did not allow time for tax shields to be fully utilized.(9) To estimate the ability to utilize all tax shields--even with the eight-year horizon -- the earnings growth required to use all tax shields within eight years was calculated. Exhibit 4 shows a need for significant improvement in order to use the increased tax shields. For those firms not using all the increased tax shields, average earnings growth would have to more than double and almost triple the average forecast growth rate in order to use all tax shields in eight years. The TRA '86 levy of a capital gains tax on the target firm in the case of a Section 338 election (step-up of the assets basis) appears to have effectively eliminated the step-up option, unless there is a method of avoiding this capital gains tax as the taxation literature hints. Under the TRA '86 (with step-up) tax structure, a leveraged buyout caused an increased in taxes for both a five-year and eight-year horizon. In spite of all the changes, a significant portion of buyout premiums still appear to be caused by a reduction in taxes due to the additional shields created by the leveraged buyout, just as the pre-TRA '86 studies showed. It is clear that under current tax law, appropriately structured leveraged buyouts still create significant tax incentives. But, on average, after TRA '86, less than half the premium can be attributed to the reduction in taxes. (1)TRA '86 reduced the maximum corporate income tax rate from 46% to 34%, thus reducing the benefits of tax shields. (2)Data for this study were obtained from various sources. Basic financial data were obtained from Value Line Disclosure database and Moody's Industrial Manual; the identification of each LBO and the description of each final LBO acquisition bid are from Mergers & Acquisitions; the earnings growth forecasts are from Value Line Surveys; the equity market prices two months prior to the initial bid announcements are from the Wall Street Journal; the initial bid announcement for each LBO is from the Wall Street Journal Index; and the LBO debt interest rates are from Securities & Exchange Commission filings. (3)Our estimates use the actual interest rates paid by each company and assume a 90% debt ratio. The results of our analysis do not change appreciably for any reasonable change in the debt ratio. (4)Managers typically had about a five-year horizon for cashing out of a leveraged buyout. This horizon seems to be lengthening in recent years (see Anders [1], and Jensen, Kaplan and Stiglin [11]). (5)The before-tax cost of debt is used since debt is similar to equity for LBOs that typically finance about 90% of assets with debt. Also, Exhibit 4 shows that tax shields will not be fully utilized, causing the after-tax cost of debt of equal the before-tax cost of debt. Use of the before-tax cost of debt of discount tax shields is consistent with Kaplan [14, p. 618]. (6)Recapture is estimated separately for each company. The depreciation rate is estimated from published data and compared with the equivalent rate under straight-line depreciation. This difference gives a positive or zero "recapture rate," which is then applied to the value of depreciable assets at the time of the leveraged buyout. (7)Kaplan [14] found evidence that stock prices begin to rise two months prior to a bid. (8)Many leveraged buyouts have divested assets at a gain. We have not tracked the tax impact of these events due to their complexity and lack of adequate data. (9)Much of the difference between pre-TRA '86 and TRA '86 structures is caused by the higher effective tax rate (46%) under pre-TRA '86.

References

[1]G. Anders, "RJR Swallows Hard, Offers $5-A-Share Stock," Wall Street Journal (December 18, 1990), p. C1. [2]G.P. Baker and K.H. Wruck, "Organizational Changes and Value Creation in Leveraged Buyouts: The Case of the O.M. Scott & Sons Company," Journal of Financial Economics (December 1989), pp. 163-190. [3]B.I. Bittker and J.S. Eustice, Federal Income Taxation of Corporations Shareholders, Boston-NY, Warren, Gorham and Lamont, 5th edition, 1987. [4]M.M. Blair, "A Surprising Culprit Behind the Rush to Leverage," Brookings Review (Winter 1989-1990), pp. 19-26. [5]G. Brode, Jr., "General Utilities Repeal: A Transactional Analysis," Journal of Taxation (June 1987), pp. 322-328. [6]J.N. Bush and M.D. Mullaney, "Basic Allocation for a Target's Assets Under the New Section 338 Temporary Regulations," Journal of Taxation (June 1986), pp. 328-333. [7]General Utilities & Operating Co., 296, U.S. 200, 1935. [8]C. Hayne, "Tax Attributes as Determinants of Shareholders Gains in Corporate Acquisitions," Journal of Financial Economics (June 1989), pp. 121-153. [9]G.A. Jarrell, J.A. Brickley, and J.M. Netter, "The Market for Corporate Control: The Empirical Evidence Since 1980," Journal of Economic Perspectives (Winter 1988), pp. 49-68. [10]M.C. Jensen, "Takeovers: Their Causes and Consequences," Journal of Economic Perspective (Winter 1988), pp. 21-48. [11]M.C. Jensen, S. Kaplan, and L. Stiglin, "Effects of LBOs on Tax Revenues of the U.S. Treasury," Tax Notes (February 6, 1989), pp. 727-733. [12]D.W. Joy and R.L. Hahn, "How to Compute Elective Formula Under Section 338 Where There is Recapture," Journal of Taxation (June 1986), pp. 334-337. [13]S. Kaplan, "Campeau's Acquisition of Federated: Value Destroyed or Value Added," Journal of Financial Economics (December 1989), pp. 191-212. [14]S. Kaplan, "Management Buyouts: Evidence on Taxes as a Source of Value," Journal of Finance (July 1989), pp. 611-632. [15]M.J. Kliegman, "Do Mirror Transactions Survive the 1986 Act?," Journal of Taxation (April 1987), pp. 206-209. [16]S.T. Limberg, "Master Limited Partnerships Offer Significant Benefits," Journal of Taxation (August 1986), pp. 84-92. [17]J.M. Lynch, P.M. Baldasaro, and N.S. Siegel, "Strategies to Maximize Benefits in Leveraged Buyouts after TRA '86," Taxation for Accountants (November 1988), pp. 304-309. [18]Mergers and Acquisitions (May-June 1987), p. 70. [19]Mergers and Acquisitions (May-June 1991). [20]W.P. Osterberg, "LBOs and Conflicts of Interest," Economic Commentary, Federal Reserve Bank of Cleveland (August 15, 1989), pp. 1-5. [21]M.S. Scholes and M.A. Wolfson, "The Effects of Changes in Tax Laws on Corporate Reorganization Activity," Journal of Business (January 1990), pp. S141-S164. [22]A. Shleifer and R.W. Vishny, "Value Maximization and the Acquisition Process," Journal of Economic Perspectives (Winter 1988), pp. 7-20.