Acquiring Intangible Assets

Transcription

1 Acquiring Intangible Assets Intangible assets are important for corporations and their owners. The book value of intangible assets as a percentage of total assets for all COMPUSTAT firms grew from 6% in 99 to % in 4. This book value calculation is even an understatement, because assets on the balance sheet almost always excludes intangible assets created within firms. Based on estimated values, Corrado and Hulton () and Peters and Taylor (5) report that intangible assets constitute 4% to 45% of total assets. For the unavailability of data on intangible assets, there is however little understanding on how they create value to shareholders. From June, Statement of Financial Accounting Standards No. 4 requires that an acquirer allocates purchase price to the fair value of identifiable intangible assets of the target (FASB, ), allowing us to evaluate the importance of intangible assets in a merger and acquisition (M&A) context. Because acquiring firms report the fair value based on their expected cash flow generated from intangibles, this measure provides an unbiased expectation. We use data generated by this reporting requirement to answer such important questions as: Why do firms acquire intangible assets? Do such acquisitions create value? What is the exact channel of value creation? Two commonly cited motivations of M&A activities focus on the creation and destruction of shareholder wealth. We apply them to study the acquisition of intangibles. The neoclassical view argues that firms purchase assets efficiently across industries they operate, so M&A activities benefit shareholders (Maksimovic and Phillips, ). The other influential view in the literature is that entrenched managers undertake M&A activities that reduce shareholder wealth (Jensen, 986; Morck, Shleifer, and Vishny, 99). Although there is a large literature supporting both views, almost all papers consider acquired assets homogeneously. Said differently, the valuation effects of acquiring tangible assets are well known, but the effects of acquiring intangible assets are still an unresolved question. It also remains an unresolved question as to what creates value. Under the neoclassical view, shareholder value is created either through wealth transfers from other stakeholders or through economies of scale, synergistic gains, and better managerial skills. Kim and Singal (99) analyze the airline mergers and find that merged firms immediately gain from less competitive pricing, which is consistent with the increased market power hypothesis. Focarelli and Panetta () emphasize analysis on longer periods to test the efficiency hypothesis, because realizing improvements in efficiency takes time. They find that efficiency gains dominate over the market power effect in the long run. Analyzing product quality and price of merged firms, Sheen (4) also documents that operational efficiencies materialize in two to three years. Considering a long-term horizon, Healy, Palepu, and Ruback (99) similarly document post-merger

2 improvements in operational efficiency. On the entrenchment view, Harford, Humphery-Jenner, and Powell () find that managers choose low synergy targets and avoid all-equity payments when acquiring private firms or public firms with blockholders. This paper evaluates the above hypotheses in relation to intangible assets. After, firms report significant prices to purchase intangibles. For example, Procter & Gamble paid nearly 5% of the deal value of $54 million to acquire Gillette s brand in 5. Using hand-collected data on,48 mergers and acquisitions between and 5, we also find that a typical acquirer in our sample pays 5% of the deal value to purchase intangible assets, which includes brands, trademarks, service marks, patents, legal contracts, etc. Our analysis begins with investigating the determinants of acquiring intangibles. Modeling the likelihood of intangible assets acquisition, we find lower levels of debt, operating performance, and capital expenditures as important acquirer characteristics. However, when we model the amount of intangible assets as a percentage of deal value, we find that acquirers of intangibles are smaller in size and incur less capital expenditures, but spend more on R&D expenses. Overall, the finding that intangible asset acquirers allocate significantly lower amount of resources on capital expenditures leading to acquisitions resonates across models. Along with our later finding that these acquirers continue reducing capital expenditures after the acquisition, this result suggests the tightening of acquirer industry. We next explore market reactions of intangible asset acquisition. While the five day cumulative abnormal return (CAR) of the whole sample is.85%, there is significant variability across acquisition of intangible assets. Specifically, firms that do not acquire intangibles experience only.7% abnormal return. The acquirer CAR then monotonically increases across terciles, ending with the highest CAR of.56% (pvalue =.) for the third tercile of firms acquiring the highest amount of intangible assets. The multivariate regression analysis that controls for a host of control variables also documents a similar result. Specifically, we find that a % increase in the acquisition of intangible assets is associated with a.7% higher CAR, which is significant at the % level. Moreover, the effect is significantly more pronounced if the target is a public firm, suggesting that acquirer shareholders benefit more from the management of intangible assets of public targets. Overall, we document that acquiring intangible assets monotonically increases shareholder wealth. Our subsequent tests analyze the sources of value creation. To evaluate the effect of intangible asset acquisition to subsequent performance, we categorize acquirers at the time of M&A announcements. We create terciles of acquirers based on their purchase of intangibles and track their performance after acquisitions. We also assign firms that do not acquire any intangible assets to a different portfolio and use

3 it as a cross-sectional benchmark. We then calculate different ratios based on acquirer three year average performances before and after acquisition to measure the effect of intangible asset acquisitions. Beginning with return on asset (ROA), we find that firms not acquiring intangibles experience increased performance that is not statistically significant. However, post-merger firm ROA improves monotonically with higher amount of intangible asset acquisition. We document the highest ROA increase of 8.8% for the third tercile, which is significant at the 5% level. To further investigate the underlying mechanism of the increased ROA, we calculate its two components: pre-tax operating margin and asset turnover. The pre-tax operating margin measures profits on a dollar of sales and the asset turnover ratio indicates operational efficiency. We again do not find any significant changes for acquirers that do not purchase intangible assets. However, we find significant improvements in the pre-tax operating margin that increases with more acquisition of intangible assets. The asset turnover ratio reduces for all acquisitions of intangible assets. These results do not support Healy, Palepu, and Ruback (99) s finding on the efficiency gains after M&A activity. We also find that firms acquiring the most intangible assets reduce R&D and capital expenditures drastically: R&D expenses reduce from 9% before the year of acquisition to 4.% after ten years, whereas capital expenditures drop from 7% to.7%. During the same event time period, the average acquirer in the third tercile increases its market share from 8.% to 4.% while the number of firms in the acquirer industry drops by 6%. These results suggest that acquirers of intangible assets benefit from

4 Table : Sample Distribution by announcement year and intangible assets acquisition. The table describes our sample of merges and acquisitions from to 5 listed on SDC where U.S. public acquirer who discloses offer price allocation gains control of a public, private, or subsidiary target whose transaction value is at least $ million and % of the acquirer s market value. Panel A: Announcement Year Number of deals Acquirer market value ($mil) Deal value ($mil) Acquired Intangible Assets ($mil) Acquired Intangible Assets / Deal value (%) 5-days CAR (%) 95, , , , Overall 48,

5 Table : Acquirer, target, and deal characteristics. The sample consists of 48 completed U.S. mergers and acquisitions listed on SDC from to 5 made by acquirers who disclose offer price allocations. Panel A: Whole Sample Mean St Dev Q Median Q Acquirer characteristics Market value of equity ($mil),75 8, ,455 Market value of assets ($mil) 4,478 5,85 77,8 Leverage Tobin s q Free cash flow Stock price run-up Deal characteristics Transaction value 4, Relative size Competed deals (dummy) Cash in payment (%) Pure cash deals (%) Pure equity deals (%) Tender-offers (%) Hostile Public target (%) Private target (%) Subsidiary target (%) High tech (dummy) Diversifying (dummy) Industry M&A Panel B: Sample with intangible assets Mean St Dev Q Median Q Acquirer characteristics Market value of equity ($mil),74 8, ,445 Market value of assets ($mil) 4,56 5,466 69,5 Leverage.76*** Tobin s q.55*** Free cash flow Stock price run-up Deal characteristics Transaction value 4, Relative size.96*** Competed deals (dummy) Cash in payment (%) Pure cash deals (%) Pure equity deals (%) Tender-offers (%)

6 Hostile Public target (%) Private target (%).5*.5... Subsidiary target (%).57*** High tech (dummy).*** Diversifying (dummy) Industry M&A Panel C: Sample without intangible assets Mean St Dev Q Median Q Acquirer characteristics Market value of equity ($mil),95 9, ,644 Market value of assets ($mil),4, ,8 Leverage Tobin s q Free cash flow Stock price run-up Deal characteristics Transaction value 448,5 5 7 Relative size Competed deals (dummy) Cash in payment (%) Pure cash deals (%) Pure equity deals (%) Tender-offers (%) Hostile Public target (%) Private target (%) Subsidiary target (%) High tech (dummy) Diversifying (dummy) Industry M&A

7 Table : Probit regressions describing what kind of firm wants to acquire intangible assets. In Probit regression, the dependent variable is a dummy variable that equals if offer price allocated to intangible assets is positive, and if zero. In OLS and Tobit regression, the dependent variables are the ratio of acquired intangible assets to transaction value. The independent variables are variables representing acquirer characteristics, which are defined in Appendix. Heteroscedasticity-robust t-statistics are reported in parentheses. *, **, *** denote significance at the, 5, and percent levels, respectively, for a twotailed test. Probit Tobit Intercept.7***.575*** (.98) (4.9) Log(MA).5 -.9** (.) (-.6) Tobin s q..5 (.) (.6) Free Cash Flow (.7) (-.4) Leverage -.5*** -.7 (-.6) (-.57) ROA -.86** -.5 (-.4) (-.9) R&D.5.6** (.84) (.44) Capital Expenditure -.8** -.5** (-.) (-.98) HHI (-.) (-.88) Number of Observations Pseudo (Adjusted) R Year Fixed Effects X X Year * Industry Fixed Effects X X 7

8 Table 4: Regression analysis of announcement abnormal returns. The dependent variable is the bidder s five day cumulative abnormal return measured using the market model. The independent variables are variables representing acquired intangible assets, product market completion, acquirer characteristics, and deal characteristics, which are defined in Appendix. Heteroskedasticity-robust t-statistics are reported in parentheses. *, **, *** denote significance at the, 5, and percent levels, respectively, for a two-tailed test. Panel A: Acquirers 5-days CAR CAR (%) p-value N without intangible assets.7.88 with intangible assets Low High Panel B: Regression Results Variables Intercept (-.8) (.) Intangible/transaction_value.7*** (.76) Intangible/transaction_value * Public 5.7** (.96) Intangible/transaction_value * Non-public.8*** (.8) Competitive Industry.6. (.54) (.55) Unique Industry (-.) (-.7) Log(MA) -.4*** -.*** (-.58) (-.48) Tobin s q.6.56 (.6) (.) Free Cash Flow (-.6) (-.58) Leverage.98**.757** (.) (.8) Stock Price Runup -.5*** -.54*** (-.) (-.5) Industry M&A (.99) (.8) Relative Size.4***.7*** (.9) (.) High Tech (.9) (.98) High Tech * Relative Size *** -9.7*** (-.7) (-.7) Diversifying (-.8) (-.6) Tender Offer.45. (.) (.) Competed (.) (.5) 8

9 Public -.846*** --.46*** (-4.84) (-4.79) All Cash..9 (.5) (.44) All Equity (-.6) (-.6) Number of Observations Adjusted R.4.6 Year Fixed Effects X X 9

10 Table 5: Operating performance and investment policy ROA Pre-tax operating margin Asset turnover %intangible Difference N Difference N Premerge Postmerge Premerge Postmerge Premerge Postmerge Difference N without intangible Low ** ** ** *** 9 High ** ** *** 9 R&D rate Capital Expenditure rate Non-debt tax shield Difference N Difference N Premerge Postmerge Premerge Postmerge Premerge Postmerge Difference N without intangible Low * *** *** *** High ** *** ** 5

11 Table 6: Information Asymmetry whole without intangible Low Median High Number of estimate (8) (7) (8) (8) (7) Forecast dispersion (.75) (.8) (.98) (.49) (.65) N Equity only (%) (4.8) (4.4) (4.) (4.) (8.9) Cash only (%) (49.) (49.) (47.7) (49.) (5.) Cash/deal (%) (4.9) (4.9) (4.7) (4.) (49.) Equity/deal (%) (4.9) (4.9) (4.7) (4.) (49.) Completion days (7) (75) (84) (6) (6) N

12 Figure.Long-term activities Market Share # firms R&D Capex

13 NDTS ROA Healy turnover

14 Appendix. Variable Definitions CAR(-,+) Log(MA) Tobin s q Leverage Free cash flow Stock price runup Pre-tax profit margin ROA R&D Capital expenditure HHI Public Private All cash All equity Diversifying acquisition Relative size High tech Industry M&A Tender offer competed Competitive industry Unique industry Five-day cumulative abnormal return (in percentage points) calculated using the market model. The market model parameters are estimated over the period (, ) with the CRSP equally-weighted return as the market index. Log(bookvalueoftotalassets + Fiscal-year-end market value of equity book value of equity). (bookvalueoftotalassets + Fiscal-year-end market value of equity book value of equity)/ book value of total assets. (Long-termdebt + debt incurrentliabilities)/ book value of total assets. (Operating income before depreciation interest expenses income taxes capital expenditures)/ book value of total assets. Bidder s buy-and-hold abnormal return (BHAR) during the period (, ). The market index is the CRSP value-weighted return. (EBIT + depreciation)/ sales. sales/market value of assets at the beginning of year (bookvalueoftotalassets + Fiscal-year-end market value of equity book value of equity). Research expenditures/ book value of assets at the beginning of year. Capital expenditures/ book value of assets at the beginning of year. Dummy variable: for public targets, otherwise. Dummy variable: for private and subsidiary targets, otherwise. Dummy variable: for purely cash-financed deals, otherwise. Dummy variable: for purely equity-financed deals, otherwise. Dummy variable: if bidder and target do not share a Fama French industry, otherwise. Transaction value (from SDC)/acquirer market value of equity (Number of shares outstanding * the stock price at the th trading day prior to announcement date). Dummy variable: if bidder and target are both from high tech industries defined by Loughran and Ritter (4), otherwise. The value of all corporate control transactions from SDC for each prior year and Fama French industry divided by the total book value of assets of all Compustat firms in the same Fama French industry and year. Dummy variable: for tender offer, otherwise. Dummy variable: if there are multiple acquirers according to SDC, otherwise. Dummy variable: if the bidder s industry is in the bottom quartile of allfama-french 48 industries' Herfindal-Hirschman Index in the year prior to announcement, otherwise. (calculated as the sum of squared market shares of all Compustat firms in each industry ). Dummy variable: if the bidder s industry is in the top quartile of allfama French 48 industries annually sorted by industrymedian product uniqueness, otherwise. (calculated as selling expense scaled by sales). 4

15 5