Excess cash flows and diversification discount

Financial Management (Financial Management Association), Summer, 2004 by John A. Doukas, Ozgur B. Kan

We study the impact of diversification on firm cash flows and excess value. Specifically, we investigate whether there is a direct link between the discount to diversification and excess cash flow reductions around related and unrelated acquisitions. Our results provide empirical support for a positive and significant association between excess cash flow declines and excess value losses after the acquisition. Our findings also show that bidders who conduct unrelated acquisitions experience larger excess cash flow declines and valuation discounts than do bidders who engage in related acquisitions. Our results are robust to the targets' excess cash flow and valuation characteristics.

**********

Recent empirical studies find that the average diversified firm trades at a discount relative to a portfolio of comparable single-segment firms. This literature suggests that diversification itself might be the reason that diversified firms produce cash flows different from those they would generate if broken into single-segment firms operating in the same line of business with diversified firm's distinct lines of business (see Berger and Ofek, 1995 among others). (1) The possibility that conglomeration might reduce cash flows motivates this article. However, several papers (Campa and Kedia, 2002; Chevalier, 2000; Graham, Lemmon, and Wolf, 2002; Whited, 2001; and Villalonga, 2004a, 2004b) in one way or another are skeptical about the diversification discount. Moreover, Mansi and Reeb (2002) show that the measure of excess value, as developed by Lang and Stulz (1994), and Berger and Ofek (1995), creates a downward bias in diversified firms, because it captures only the shareholder value of the firm. They argue that diversification reduces firm risk and, therefore, it does not have adverse bearing on the cash flows of the firm. However, they do not provide evidence in support of this conjecture. While the Lamont and Polk (2001) study, based on aggregate data, shows that firm value is inversely related with subsequent returns and that the diversification discount manifests itself in both expected returns and expected cash flows, it does not provide direct evidence in support of a link between firm cash flows and excess value.

If the value of the firm is the discounted value of its cash flows and if there is a discount to diversification, then the discount should be either because cash flows decline, or the discount rate rises with diversification. In this article, we examine the impact of diversification on firm cash flows and excess values and investigate whether there is a direct association between changes in cash flows and the diversification discount (i.e., excess value) around the act of diversification, (2) We examine this relation for multi-segment firms that engage in related and unrelated acquisition transactions and compare their differences. We also analyze the impact of targets on the bidders' excess cash flows and diversification discount. To address these issues, we use a sample of 742 firm-year acquisitions completed by US firms over the 1991-1997 period.

Our results provide empirical support for a positive association between reductions in cash flows and changes in the diversification discount subsequent to acquisitions. Our findings show that a 10% decline in bidders' excess cash flows is associated with a 3.7% excess value loss. Second, we find that bidders that acquire unrelated targets experience greater excess cash flow declines and valuation discounts than do bidders involved in related acquisitions. Third, our results are robust to the targets' excess cash flow and valuation characteristics. Finally, our evidence indicates that industrial diversification is driven by bidders' industry growth opportunities.

The article proceeds as follows. Section I describes our data sources, sample selection procedure, and sample characteristics. Section II presents the pre-acquisition excess cash flow and valuation performance of bidders. In Section III, we assess the change in the excess performance of multi-segment bidders involved in related and unrelated acquisitions in subsequent years, and examine the relation between changes in bidders' excess value and excess cash flows. Section IV examines the impact of targets' characteristics on bidders' post-acquisition excess cash flows and valuation. Section V concludes.

I. Data Sources, Sample Selection, and Industrial Classification of Bidders

We begin by describing the sources of data, sample selection procedure, industrial classification of bidders, and sample characteristics.

A. Sources of Data and Sample Selection

Our sample consists of domestic acquisitions conducted by the US bidders between January 1, 1991 and December 31, 1997 as reported in the Domestic Acquisitions roster of Securities Data Corporation's Mergers & Acquisitions (M&A) Journal. The rosters of the M&A Journal include all acquisitions of $5 million value or higher and report the name of the target, the Standard Industrial Classification (SIC) code (at the 2-digit level before 1993, at the 4-digit level starting 1993), the business definition of target firms or businesses, the name, and the business definition of bidder firms. The rosters also report the value of acquisition, the method of payment, whether the target is divested or not, the completion day of the acquisition, and the advisors to both parties. We exclude from the sample all transactions associated with target firms in non-manufacturing industries (i.e., Finance, Insurance and Real Estate, with 2-digit SIC codes from 60 to 67, and Services, with 2-digit SIC codes from 70 to 89). We also exclude bidders involved in both domestic and cross-border acquisitions in the same calendar year. Our initial sample covers 10,128 domestic acquisitions over the 1991-1997 period. We also use Compact Disclosure/Worldscope Global CDs to obtain insider and institutional ownership information for the sample firms.