# Do family owners use firm hedging policy to hedge personal undiversified wealth risk?

We examine whether family ownership affects the value impact of the operational and financial dimensions of firms 'hedging policies. We show that family firms' market valuations are higher than those of non-family firms, consistent with the view that family firms benefit from family owners' long-term perspectives and ability to monitor managers. In addition, while both operational and financial hedging policies per se are valuable in non-family firms, they do not create any value in family firms. These results support the notion that the founding families' need to hedge the risk of their undiversified personal wealth portfolio leads to suboptimal risk management decisions.**********

Family ownership is quite common and significant in US firms across a broad range of industries. Firms with founding families' presence constitute about one-third of the S&P 500 and Fortune 500 industrial firms, and founding families represent the most common types of large, undiversified shareholders (Shleifer and Vishny, 1986; Villalonga and Amit, 2006). As reported in Anderson and Reeb (2004), excluding 97 banks and public utilities, founding families are present in 141 of 403 S&P 500 firms and hold, on average, about 17.9% of equity stakes and 20% of the board seats in these firms.

Prior literature has provided mixed evidence on the benefits firms derive from substantial ownership by founder families (hereafter "family firms"). (1) Based on Fama and Jensen (1983), concentrated shareholdings allow for an exchange of profits for private rents. Hence, it is possible that combining ownership and control in family firms may result in a suboptimal ownership structure. Agency problems (Demsetz, 1983; Shleifer and Vishny, 1997) between controlling and noncontrolling shareholders are more likely to arise within family-owned firms (Anderson and Reeb, 2004; Anderson, Duru, and Reeb, 2009).

However, family ownership may be beneficial to outside shareholders. For instance, a founding family presence facilitates more efficient monitoring of firm managers (Demsetz and Lehn, 1985). In addition, family owners' long-term horizons and their preference for long-term investments may mitigate managerial incentives for myopic investment decisions, thereby leading to greater investment efficiency (James, 1999). Anderson and Reeb (2003a,b) show that US family firms exhibit better accounting and market measures of performance than non-family firms.

Our study contributes to the literature by providing evidence on the role of family owners in a specific type of corporate decision, the risk management policy of the firm. (2) We examine the managerial hedging decisions of family firms and their impact on firm value to assess whether family ownership is associated with less severe agency costs than non-family firms. In particular, we compare family-owned and non-family-owned S&P 500 firms in terms of the value impact of industry diversification and derivatives use. These may be thought of as proxies for the operational and financial dimensions of firms' hedging (or risk management) policy, respectively. Risk management decisions have been found to either exacerbate (e.g., in the case of speculative use risk management policies, see Geczy, Minton, and Schrand, 2007; Aabo, Hansen, and Pantzalis, 2012, among others) or mitigate (DaDalt, Gay, and Nam, 2002; Lin, Pantzalis, and Park, 2007) agency problems. Hence, analyzing them may yield insights into family ownership valuation effects.

We posit that there should be differences between family and non-family-owned firms in terms of the impact of hedging policies on firm valuation. Family ownership, usually in the form of large, undiversified blockholdings, may increase the propensity to use hedging policies to reduce personal wealth portfolio risk at the expense of firm value. Specifically, if family owners seek to reduce firm-specific risk in order to hedge their undiversified wealth portfolio, they may try to do so by influencing firm hedging policy through the use of derivatives in the short run and the adjustment of operating policies in the long run. Alternatively, since hedging policy can improve firms' information environments by reducing information asymmetry between managers and the financial markets, family firms that are often regarded as more opaque (Anderson et al., 2009) may use hedging policies to reduce information asymmetries.

Our empirical investigation is performed in three stages. We start by examining how family ownership affects the value of operational hedging. Next, we investigate the effect of financial hedging and finally examine both dimensions of hedging policy in combination. Our univariate results on operational hedging show that family firms exhibit higher values and larger diversification discounts than non-family firms. Multivariate tests also show a negative effect of diversification on valuation (as in Berger and Ofek, 1995, among others) and a larger diversification discount for family firms than for non-family firms. Interestingly, after controlling for the endogenous relationship between diversification and firm value, we find that the average firm experiences a value premium through diversification. This result holds for alternative valuation measures, as well as alternative estimation methods. Moreover, the diversification premium disappears in the case of family firms. Therefore, our results are consistent with diversification strategies being used by undiversified family owners of family firms to diversify their personal wealth portfolios.

In the second stage of our empirical examination, we show that financial hedging policy is also positively associated with firm value. This effect is stronger for firms with policies relying on a greater number of different financial derivatives contracts. It indicates that comprehensive and sophisticated financial risk management policies tend to have significantly positive value impacts (Lin, Pantzalis, and Park, 2009). However, we also find that the value impact of financial hedging policy is not significant for family firms, implying that as in the case of operational hedging policies, family owners may use financial hedging policies to reduce the risk of their own undiversified wealth portfolio, thereby reducing their effectiveness.

Moreover, when we examine the value impact of financial hedging conditional upon corporate diversification (i.e., in the presence of operational hedging capability), we find that the use of derivatives adds value for firms that use them regardless of whether the firms are family owned or not. This finding implies that when firms are diversified across many industries, family owners do not need to use financial hedging by the firm to reduce their own undiversified wealth risk. Therefore, the benefits from derivatives use are preserved. This finding is in line with Lin et al. (2007) who note that diversified firms benefit from financial risk management as derivative usage lowers information asymmetry, thereby reducing the negative valuation effects of diversification.

Overall, the story that emerges from our findings is one of family owners whose familiarity and/or superior information about their firms' core business allows them to earn higher returns in their investments in the family firm. Consequently, family owners who typically prefer control do not diversify their firms into other industries. Moreover, family owners of focused firms should be more knowledgeable about the business risk in the firms' core industry and less reliant on financial hedges as a risk management tool.

However, block investments in focused firms also imply greater exposure to undiversified wealth risk. Thus, for family owners of focused firms, the operational hedging dimension of risk management is often not available and the risk of the focused firms' assets is greater. Therefore family owners tend to derivatives to hedge their own undiversified personal wealth.

Family owners' need to rely on financial hedging policy to hedge undiversified wealth diminishes when the firm is diversified, as both the overall firm risk and the cost of operational hedges are lower. Thus, family ownership does not diminish the complementarities that exist between operational and financial hedging dimensions of risk management policies in diversified firms.

Collectively, we interpret these results as consistent with the view that family owners are effective in monitoring and have longer investment horizons that lead to greater investment efficiency. However, our evidence also suggests that risk management decisions in family firms do not necessarily lower overall agency problems and enhance firm value. In fact, family firms seem to suffer greater agency problems when they attempt to diversify out of core business and especially when they use redundant derivatives contracts. Anderson et al. (2009) find that family ownership can improve value in firms with transparent information environments, while it can destroy value in firms with opaque information environments. Allayannis and Schill (2010) argue that rogue policies are indicative of agency problems leading to valuation discounts. In sum, our evidence supports the notion that family firms' attempts to establish operational or financial hedges may be motivated by the self-interest of the family owners to diversify their personal wealth portfolio. This study contributes to the literature by shedding light on the role of family owners in firms' risk management policy. In doing so, we provide a partial explanation for the reason why previous studies have found mixed evidence regarding the relationship between firm performance and family ownership.

Our study differs from Allayannis and Schill (2010) in that we do not examine how other financial policies jointly affect risk management and valuation, but rather focus on the role of founding families in the relationship between risk management policy and valuation. The narrower focus on founding families allows for more direct evidence that sheds light on a controversial issue (i.e., whether and when do firms benefit or suffer from the presence of family owners).

The paper is structured as follows. The next section relates this study to the existing literature and provides predictions, followed by the description of data sources and the construction of excess value. Sections III to VI discuss the methodology and present the empirical results concerning the impact of family ownership on the correlation between the operational and financial aspects of risk management to firm value. Section VII provides various robustness tests. The last section summarizes our study and presents our conclusions.

I. Related Literature

A. Family Firms and Valuation

Large, publicly traded family firms should benefit from a greater combination of ownership and control, thus alleviating classic owner-manager conflicts. Demsetz and Lehn (1985) argue that large shareholders typically act to mitigate managerial expropriation. Founding families usually maintain large, undiversified equity positions in their firms and their experience with the firm allows them to act as effective monitors with strong influence on management. In addition, as James (1999) posits, founding families' shareholders have longer investment horizons than other shareholders and their presence constitutes an influence for firms toward greater investment efficiency. Previous literature has positively linked the long investment horizons of management to firm valuation (Stein, 1988, 1989; Antia, Pantzalis, and Park, 2010).

However, family firms' concentrated ownership may be inherently less efficient leading to greater agency costs. For example, past studies have demonstrated that concentrated shareholdings are associated with an exchange of profits for private rents (Fama and Jensen, 1983), nonpecuniary consumption, or a tendency to extract private benefits (Shleifer and Vishny, 1997). Anderson and Reeb (2004) document that moderate family ownership generally provides benefits to outside shareholders, but in the presence of relatively few independent directors, family ownership has a negative impact on firm performance. In a more recent paper, Anderson et al. (2009) find that family firms are more opaque than firms with a diffuse shareholder base. Since corporate transparency is important in protecting minority investors' rights against controlling shareholders, it follows that family ownership and control can lead to greater agency problems and a negative impact on firm value.

B. Corporate Hedging Policy and Valuation

As previously mentioned, we proxy firms' ability to construct operational hedges by the degree of their industry diversification, which many past studies in the financial economics literature have found to be negatively related to firm value (Lang and Stulz, 1994; Berger and Ofek, 1995; Servaes, 1996; Wernerfelt and Montgomery, 1988; Lins and Servaes, 1999). (3) However, in recent years, the findings in earlier studies regarding diversification discounts have been challenged on the grounds of endogeneity in the relationship between valuation and the decision to diversify (Campa and Kedia, 2002; Villalonga, 2004a) and of the possibility of measurement errors (Whited, 2001; Graham, Lemmon, and Wolf, 2002; Mansi and Reeb, 2002; Villalonga, 2004b). Accordingly, in this study we account for the endogenous nature of diversification decisions following Campa and Kedia (2002), and also use several alternative measures of excess valuation. Operational hedging policy has been shown to be effective in reducing firms' exposure to risk (Pantzalis, Simkins, and Laux, 2001; Aabo and Pantzalis, 2011).

Financial hedging has also been shown to enhance firm value (Allayannis and Weston, 2001; Graham and Rogers, 2002; Bartram, 2004; Carter, Rodgers, and Simkins, 2006). This literature emphasizes the role of financial hedging in reducing the level of agency costs that arise when managers pursue personal goals. Stulz (1984) argues that corporate hedging decisions may be the outcome of managerial risk aversion. Outside shareholders, who can easily diversify the portfolios of their own accounts, are indifferent with respect to the amount of hedging activity that the firm undertakes. However, this is not the case for managers who typically hold a large portion of their wealth in a firm's stock. A suitable way for these managers to reduce the variance of firm value is by using hedging instruments that may make managers better off without costing outside shareholders much.

Since agency conflicts tend to be exacerbated in the presence of "information asymmetry" where managers possess superior information than that of owners, hedging may be beneficial as it has the potential to reduce such information asymmetry (DeMarzo and Duffie, 1995; Breeden and Viswanathan, 1998; DaDalt et al., 2002). Therefore, hedging policies may ease the task of forecasting future cash flow and improve stock price informativeness. DeMarzo and Duffie (1995) assert that outsiders encounter difficulty interpreting the impact of risks that corporations can hedge away. Dolde and Mishra (2007) determine that geographically diversified firms use substantially greater amounts of foreign exchange derivatives than purely domestic firms. DaDalt et al. (2002) present evidence supporting the premise that the use of derivatives is associated with lower information asymmetry. (4)

Our paper is closest to Anderson and Reeb (2003b) with regard to family ownership's impact on the relationship between diversification and valuation. They provide evidence that family firms are less likely to engage in corporate diversification, in contrast to the expectation that families have strong incentives to minimize firm risk due to the undiversified nature of their holdings and their desire for firm survival. These authors also determine that family firms have lower values when they are diversified. This paper extends Anderson and Reeb (2003b) by providing a broad scope of tests on financial diversification strategies, in addition to operational diversification strategies and their effects on firm value.

II. Sample Selection and the Construction of Excess Value

A. Data Sources

We obtain family firm information from David Reeb's Web site (http://astro.temple.edu/~dreeb/ Working2.html). This data was used in Anderson and Reeb (2004) where they studied the effect of founding family ownership on firm performance in the context of agency theory. In that paper, they use the 1992 Standard & Poor's (S&P) 500 firms from 1992 to 1999 and find that founders control about one-third of the S&P 500 and Fortune 500 industrial firms, representing the most persistent and common types of large, undiversified shareholders in the US market. To construct the data on founding family ownership, these authors manually collected information from corporate proxy statements. The data includes 403 nonbanking/nonutility firms and 2,686 firm-year observations.

Accounting and financial data are drawn from Compustat. Firms with a market value of equity less than $20 million are excluded in order to avoid cases of firms with distorted valuation multiples in the mispricing measures. As in Anderson and Reeb (2003a,b), we exclude any firms from our initial sample if they are regulated utilities (standard industrial classification [SIC] codes 4910-4949), depository institutions (SIC codes 6000-6099), and holding or other investment firms (SIC codes 6700-6799). This is because government regulations potentially influence ownership structures and because it is difficult to measure market-based performance for those firms. To construct an alternative measure of a firm's excess value (i.e., Ohlson's (1995) residual income model), we use analysts' forecasts available in the Institutional Brokers Estimate System (I/B/E/S) Detail History data set. As presented later, some economic variables are used to estimate the probability of a firm's industrial diversification (see Table III). Among them, we extract the data on mergers and acquisitions from the Securities Data Company (SDC), and gross domestic product (GDP) from the National Bureau of Economic Research (NBER). The final sample includes 2,539 firm-year observations with 383 firms during our sample period from 1992 to 1999.

We further collect information concerning the derivatives use of S&P 500 firms from the Database of Users of Derivatives published by Swaps Monitor Publications, Inc. The database compiles information regarding the use of firm derivatives according to Statement of Financial Accounting Standard (hereafter SFAS) 105, which became effective in July 1990. SFAS 105 requires firms to report information about financial instruments, such as forwards, futures, options, and swaps that have off-balance sheet risk. Subsequent to SFAS 105, two more statements, SFAS 107 (effective December 1992) and SFAS 119 (effective December 1994), were promulgated during the sample period. The database's contract spreadsheets contain information about the firms that list notional amounts of over-the-counter and exchange-traded interest rate and currency derivatives outstanding at period end.

Swaps Monitor started compiling this database in 1992 and ceased in the third quarter of 1997. Thus, our second sample, used to examine the value impact of hedging policy, is restricted to the five-year period (1992-1996) where complete annual derivatives-use data are available. To ensure that our results remain robust outside the sample period covered by the family ownership and derivatives-use data availability, we also add another, more recent year of data to our analysis. (5)

B. Measuring Firm's Excess Value

We construct the excess value measure for our sample firms by following previous studies (Doukas, Kim, and Pantzalis, 2005) that adopt the Berger and Ofek (1995) approach in order to capture the degree of drift of the current market value from the fundamental (intrinsic) firm value. Excess value is computed as:

Excess [value.sub.i,t] = Ln[[Capita.sub.i,t]/Imputed([Capital.sub.i,t])], (1)

where Capital is the total capital that is the market value of equity plus the book value of debt, Imputed(Capital) is the imputed value derived as the product of firm sales and the median capital-to-sales ratio in the firm's industry. The industry classification here is based on the Fama-French 48 sectors. (6) A positive (negative) value of excess value indicates that the market assigns value higher (lower) than what the industry benchmark-based value would indicate. Alternatively, the Excess value expression in Equation (1) can be rewritten as:

Excess [value.sub.i,t] = Ln[[(Capital/Sales).sub.i,t]]/[[(Capital/Sales).sub.m,t]]. (2)

The excess value of firm i is the natural log of the ratio of a firm's sales-adjusted market value of common equity to the industry's median (m) sales-adjusted market value of common equity.

III. Family Firms and the Determinants of Operational and Financial Hedging Policies

Table I reports descriptive statistics for the pooled sample. Panel A indicates that family firms account for 33% of the total number of observations in our sample (Family). More than 61% of firms are diversified in their operation (Diversification). In particular, 75% of the firms in our sample operate in three or fewer than three different business segments, while the average firm reports 2.34 segments (Number of Segments). In their risk management on the financial side, about 76% of the firms make derivatives contracts. Our sample firms typically use 1.1 types among interest rate (IR)- and foreign exchange (FX)-type derivatives and 1.7 different contracts, and with total expenditures of $879 million. The median market value of total assets is $4.2 billion (exponential of 22.1544), which is greater than that of other studies as our sample consists of S&P 500 firms (Size). On average, our sample firms spend 7.15% of gross sales (Capital Expenditure/Sales). Their EBIT is around 11% of sales (EBIT/Sales) and the mean long-term debt to total assets is 17.24% (Leverage). Approximately two-thirds of the firms are involved in foreign sales (Foreign Sales).

Panel B contains descriptive statistics regarding the financial hedging variables constructed from the Swaps Monitor data set. The Swaps Monitor lists the notional dollar amounts of seven different contracts spanning two general types of derivatives: 1) interest rate (hereafter IR) derivatives and 2) foreign exchange (hereafter FX) derivatives. The seven different contracts are: 1) IR options, 2) IR swaps, 3) IR forwards/futures, 4) FX options, 5) FX swaps, 6) FX futures, and 7) FX forwards. In the majority of our tests, we use the four measures of corporate hedging policy. First, we create an indicator variable that takes a value of one if the firm uses any derivatives and zero otherwise (Derivatives Use). Next, we examine users' policy characteristics. In particular, we use the number of types of contracts used (Number of Types), which takes a value of zero (i.e., nonusers), one (either of IR type or FX type contract), and two (both types). Alternatively, we count the number of different contracts (Number of Contracts), which takes a value from zero to seven. We note that the largest number of contracts is six in our sample. Finally, we measure the total dollar amount of derivatives (Contracted Dollar Amount).

Panel B documents the summary statistics of the indicator variables of derivative usage in the first set of rows, the notional dollar amount of derivatives in the second set of rows, and the ratio of the amount of the derivatives contract to total capital in the third set of rows. Approximately 64% of the firms use FX derivatives, while 47% use IR derivatives. This confirms our expectation that these are firms that are heavily dependent on their foreign operations and, as such, need for currency risk management. The average dollar amount of FX derivatives ($574 million, equivalent to 4.83% of total capital) is also larger than that of IR derivatives ($469 million, equivalent to 4.34% of total capital). Within the IR derivatives, swaps account for the majority indicating the long-term nature of IR risk management. For FX derivatives, the sample firms use 69% of money in forwards. However, FX derivative usage is spread out across the different instruments, possibly indicating a more complex/sophisticated approach in FX risk management.

Table II illustrates the way that family firms differ from non-family firms in terms of firm characteristics. It reports mean values of all of the variables used in the study for the two groups, their differences, and the corresponding t-statistics for the mean difference tests.

The value of family firms is higher than that of non-family firms. The main excess value measure, as well as three alternative measures (explained later), demonstrate higher values for family firms. The mean difference between the two groups is statistically significant at the 1% level in all measures. Family firms are less diversified than non-family firms. Diversification indicates that about half of the family firms are diversified, while two-thirds of the non-family firms have several business segments. Also, non-family firms, on average, have 2.53 segments, which is a significantly larger number than 1.97, the average number of segments of family firms. Family firms are also less likely to hedge financial risk by using derivatives. The three variables concerning derivatives policy (i.e., number of types, number of contracts, and dollar amounts) find that family firms use fewer types and contracts, and spend less. The comparison of the other variables' mean values reveals that family firms are smaller, less leveraged, have smaller capital expenditures, and focus more on the domestic market.

As addressed in Anderson and Reeb (2003b), founding families have strong incentives to minimize firm risk due to the undiversified nature of their holdings and their desire to ensure firm survival. Therefore, families are expected to seek risk reduction strategies through operational or financial hedging.

To examine the determinants of a family firm's propensity to use operational hedges (i.e., to diversify) and financial hedges, we use the Campa and Kedia (2002) probit model. In the probit model, the first part of the instrument consists of firm characteristics including Size, EBIT/Sales, Capital Expenditure/Sales, and lagged and historical values of these variables; a foreign sales indicator (Foreign Sales); and an exchange variable (New York Stock Exchange [NYSE]). The second part of instrument includes industry variables, the fraction of all firms in the industry that are conglomerates (Fraction of Conglomerates), the fraction of industry sales accounted for by conglomerates (Fraction of Conglomerates' Sales), the number of announced mergers and acquisitions in the year (Number of M&As), the US dollar value of announced mergers and acquisitions in the year (Volume of M&As), the growth rate in real GDP (GDP Growth) and a lagged rate [GDP Growth (1 lag)].

Table III reports the probit estimates for diversification and derivatives use. Due to the difficulty in interpretation, we report the marginal effect of each variable. The marginal effect allows us to interpret the coefficients easily as in typical regression models. Some firm-specific and industry variables are highly correlated with the diversification variable Diversification. As documented in other studies, the historical average variables show that firms with larger total assets, lower profitability, and lower growth opportunities are more likely to engage in diversification. The significant and positive coefficients of Fraction of Conglomerates and Fraction of Conglomerates' Sales indicate that firms' propensity to diversify is greater when they belong to industries dominated by conglomerates. Firms listed in the NYSE are more likely to diversify than firms in other exchange markets.

Contrary to the moral hazard hypothesis, which predicts a greater propensity to hedge in family firms, we find that family firms are associated with significantly less corporate diversification (i.e., the ability to construct operational hedges). They also tend to use fewer financial derivatives than non-family firms, although this correlation is not statistically significant.

IV. Family Firms and the Value Impact of Diversification

A. Based on Univariate Tests and a Simple OLS Framework

The univariate results presented in Table II provide preliminary evidence supporting our original expectation that founding family presence is positively related with firm valuation. In the following test, we attempt to determine how the value change by diversification is different for family and non-family groups. To do so, we initially classify the sample firms into four different groups in terms of family presence and diversification (i.e., 2 x 2). Then, we compare the excess values of the four groups.

Panel A of Table IV indicates that firm valuation is positively associated with family presence and is negatively related to business diversification. Accordingly, we find that nondiversified family firms have the highest mean of Excess Value (0.2001) and diversified non-family firms have the lowest mean value (-0.0720) among the four subsamples. It appears as if business diversification destroys firm value for both family and non-family groups. However, in absolute terms, the difference in Excess Value between diversified firms and nondiversified firms is much larger for family firms (-0.1101) than for non-family firms (-0.0671) suggesting that the value discount caused by diversification is more pronounced in family firms than in non-family firms.

Next, we provide multifactor regression evidence regarding the existence of a diversification discount effect and how it relates to family presence. To explain firm value, we control for the family-firm indicator, a diversification dummy, and the interaction term between the two variables. Following Berger and Ofek (1995), we also include the log of total assets {Size), profitability measured as earnings before interest, taxes, depreciations and amortization (EBIT) scaled by sales, and growth opportunity computed as capital expenditures scaled by sales. We refer to this ordinary least-square (OLS) model as the Berger and Ofek (1995) model (hereafter the BO model). (7)

The results, presented in Panel B of Table IV, confirm our earlier findings from the univariate tests. The model in Column [1] does not include the diversification variable and yields a positive relationship between family firms and valuation. The estimated coefficient of Family is 0.1942 with a f-statistic of 7.38. Diversification is found to have a negative and significant effect on the excess value in the next three regressions. More importantly, as demonstrated in Column [4], the diversification discount is much more pronounced in family firms. In particular, the coefficient of the interaction between the family firm indicator and the diversification dummy is significantly negative. In that model, the magnitude of Diversification, representing the non-family firms' diversification effect on firm value, is -0.0169 and insignificant at conventional levels. This value effect is one-third of the magnitude found in Column [3] for the whole sample (-0.0539). The remaining explanatory variables display a similar relationship with excess valuation as they did in the Berger and Ofek (1995) regressions (i.e., size, EBIT, and capital expenditures are positively associated with firm value).

Overall, the univariate and OLS regression evidence from Table IV provides support to the notion that family firms are superior in value than non-family firms, but firm value is discounted more severely when family firms engage in business diversification.

B. Controlling for the Endogenous Relation between Value and Diversification

As previously mentioned, the results of simple OLS regressions can only be regarded as preliminary evidence of the relationship between valuation and diversification as they do not control for the endogenous nature of the relationship. It is possible that we have a significant endogeneity problem here because, as we noted in the discussion of the probit model results from Table III, there are some variables that are highly correlated with a firm's diversification decision that were omitted in the BO model. In fact, Campa and Kedia (2002) and Villalonga (2004a) find a value premium from diversification after controlling for endogeneity in managers' decisions to diversify into other lines of business. Therefore, we now turn to an examination of the effects of family presence and diversification on valuation by addressing the potential problem caused by the endogenous nature of diversification through the use of a two-stage least-squares (2SLS) regression. (8) In the first step, we estimate a firm's propensity to diversify using the probit model shown in the left-hand side of Table III, where the diversification dummy (i.e., Diversification) was estimated. We exclude Family from a set of instruments, which will be interacted in the second-stage model.

Next, we use the estimated Diversification in the second-stage model as an independent variable to explain firm value. As demonstrated in Table V, the second-stage model includes some additional variables (leverage and the square of size), as well as the lagged values of the variables originally included in the BO model. (9) First, we find that Campa and Kedia (2002) model does not qualitatively change the result obtained from the BO model. Even though the coefficient of the family indicator in the Campa and Kedia (2002) model is 0.1608, slightly reduced relative to the corresponding coefficient of the BO model from Table IV that was 0.1942, it remains statistically significant at the 1% level.

More interestingly, after controlling for the endogeneity of Diversification, we find that a firm's diversification becomes positively associated with valuation. In Column [2], the estimated coefficient on Diversification is now 0.2220 with a t-statistic of 3.19. Note that it was -0.0743 in the Berger and Ofek (1995) model without controlling for endogeneity (Table IV). Campa and Kedia (2002) find that once endogeneity is accounted for, the diversification discount turns into a premium. We also note an additional interesting and important result. There is an asymmetric effect of diversification on firm value between family firms and non-family firms. Column [4] indicates that the impact of diversification of firm value of non-family firms has a magnitude of 0.5114, which is greater than the diversification effect (i.e., 0.2606) for the pooled sample reported in Column [3], However, the corresponding effect for family firms is the sum of the Diversification and the Family*Diversification coefficients, which is negative (-0.0766) and insignificant (p-value = 0.4326). To address the concern of the suitability of the relative valuation approach in constructing the dependent variable in tests of diversification's impact on valuation voiced in several prior studies (Whited, 2001; Graham et al., 2002; Mansi and Reeb, 2002; Villalonga, 2004b), we also tested the models using three alternative measures of excess value. The first alternative measure, Excess Value {TA), is constructed in a similar fashion to the previous industry benchmark-based measure (Excess Value), but uses a firm's total assets instead of gross sales in the computation of the imputed value. Therefore, Excess Value [(TA).sub.i,t] = Ln [[(Capital/Total Assets).sub.i,t]/[(Capital/Total assets).sub.m,t]]. The second measure, Excess Value (RI), is the natural log of the ratio between the stock price and its intrinsic value from Ohlson's (1995) residual income value approach. Excess Value [(RI).sub.i,t] = Ln [[Price.sub.i,t]/[Imputed(Value).sub.i,t]], where Price is the stock price at the end of June of each year from Center for Research in Security Prices (CRSP), Imputed(Value) is the intrinsic value using the residual income model (Ohlson, 1995) and the median values of analysts' forecasts issued in June, as in Frankel and Lee (1998). The third measure, Excess Value {MB), is the industry-adjusted market-to-book ratio. Excess Value [(MB).sub.i,t] = Ln [[Market-to- Book.sub.i,t]/[Market-to-Book.sub.m,t]], where [Market-to-Book.sub.m,t], is the market-to-book value of total assets for firm i and [Market-to-Book.sub.m,t], is the industry median market-to-book.

The results obtained from the models that utilize the three alternative measures of excess firm value are reported in the three columns on the right-hand side of Table V. All three regressions report the same patterns of coefficients on the family firm indicator, the diversification indicator, and their interaction term. Family presence and diversification are positively related with firm value in the two-stage model, and the effect of family firm's diversification (i.e., the interaction) is negative and significant. Therefore, the results obtained using our main valuation measure, (Excess Value), are confirmed by these alternative measures of excess firm value.

Overall, the evidence in Table V indicates that family firms that do not typically venture into lines of business other than their core suffer from a value discount when they actually do so. (10) Alternatively, from the perspective of industry diversification as a proxy for a firm's ability to engage in operational hedging, the results imply that the beneficial valuation effects of operational hedging policies vanish in the presence of founding family owners. Family owners may use the operational hedging policy in a suboptimal way (i.e., to address the need to diversify their own personal wealth portfolio).

V. Family Firms and the Value Impact of Financial Hedging Policy

From the results in the previous section, we determine that operational risk management has an asymmetric influence on firm value between family and non-family firms. In this section, we conduct tests to determine whether founding family presence affects the way financial hedging policy affects firm valuation. We begin with mean difference tests that compare groups' means of excess value after sorting firms on founding family presence and on four alternative derivatives-use measures. The results are presented in Table VI. In the last column, across all four panels (each corresponding to a different measure of financial hedging policy), we observe that family firms are, on average, valued higher than non-family firms. More interestingly, this valuation difference is generally more pronounced among firms lacking a financial hedging policy. When we compare excess value holding family owners' presence constant, we observe more noteworthy results. In Panel A, Group [1] includes users of derivatives, while Group [2] includes nonusers. The average excess value for the group consisting of family firms that use derivatives is 0.1491. Interestingly, the average excess firm value is similar even for the group of family firms that do not use any derivatives (i.e., 0.1677). A similar pattern is also noted in the first column of Panels B to D when we classify family firms based on the three alternative hedging policy variables. Thus, the means difference test suggests that the in the presence of family owners, financial hedging policy does not have a significant value impact. However, this is not true for non-family firms. As noted in the second column, among non-family firms, the average excess value of firms lacking a financial hedging policy is significantly lower than that of firms with an active financial hedging policy. This result indicates that there is value in financial hedging.

Next, we provide multifactor regression evidence on the valuation impact of derivative-use policy. To see how different aspects of a firm's financial hedging policies affect firm value, we use the four previously mentioned variables: 1) Derivatives Use, 2) Number of Types, 3) Number of Contracts, and 4) Contracted Dollar Amount. Since the latter three variables are highly skewed, we transform them by adding one and taking the natural log. Then, we add the family firm indicator and the interaction term between family and derivatives variables in addition to the same set of independent variables used in the previous tables' Excess Value models.

The results, found in Table VII, support the existing evidence that corporate hedging policy is positively associated with firm value. The use of derivatives is found to have a positive and significant effect on the excess value. The estimated coefficient of Derivatives Use in Column [1] is 0.1505 with a ^-statistic of 3.13, and its magnitude is comparable to that of the operational hedging proxy, Diversification, which was 0.2220 (see Table V, Column [2]). The coefficients on the other variables (Number of Types, Number of Contracts, and Contracted Dollar Amount) suggest that firms with derivative policies that are more comprehensive and sophisticated outperform those with less comprehensive and sophisticated policies. Remarkably, when we add Family in the regression model (see Columns [2], [5], [8], and [11]), the magnitude of the estimated coefficient of the different financial hedging policy variables is not reduced. When we also account for the interactive effects of financial hedging and founding family presence, our results indicate that the interaction terms' coefficients (Family*Derivatives) are negative and statistically significant. These negative interaction terms' coefficients indicate two important points. First, confirming the univariate test results from Table VI, the value impact of financial hedging policy is different for family versus non-family firms. For example, in Column [3], we note that the use of derivatives improves non-family firms' excess value by 0.2428. However, family firms' use of derivatives results in a value discount of 0.0324 (= 0.2428 - 0.2752). In addition, the value impact of founding families in the absence of a financial hedging policy 0.4570 is quite a bit larger than the corresponding effect in the pooled sample (0.2518) that contains both users and nonusers of derivatives. The value impact of family firms using derivatives is only 0.1818 (= 0.4570 - 0.2752). The relationships of other explanatory variables to firm value remain similar as in the previous table.

VI. Founding Family Presence and the Value Impact of the Two Dimensions of Hedging Policy

In this subsection, we examine the valuation effects of the interplay between founding family presence and both of the two dimensions of corporate hedging policy. The motivation behind this test is that we expect operational hedging policy (i.e., diversification) to affect the impact of financial hedging policy (i.e., derivatives usage) on firm value. This notion is based on the Lim and Wang (2007) argument that financial hedging can increase a firm's incentive to manage risk through diversification, and is supported by empirical evidence in Geczy, Minton, and Schrand (1997) who find that geographic diversification (measured by the foreign sales ratio) is significantly and positively related to foreign currency derivatives usage. Also, Allayannis, Ihrig, and Weston (2001) determine that the more geographically dispersed a firm is, the more likely it is to use financial derivatives. They find that operational hedging (i.e., diversification) strategies increase firm value only when used in combination with financial hedging strategies. Finally, other studies (Carter, Pantzalis, and Simkins, 2001) have demonstrated that operational and financial hedges are complementary risk management strategies. Therefore, it is appropriate to examine their value impact together in the same model.

We examine the value impact of financial and operational hedging for subgroups of firms classified based upon founding family presence and provide the regression results in Table VIII. The model we estimate is formed as follows. First, we begin by creating four subgroups after sorting firms based on family presence and diversification (i.e., Diversified Family, Undiversified Family, Diversified Non-family, and Undiversified Non-family). Among them, we designate the undiversified family firms' group as the "base" group and we omit this variable from the model. Next, we include the remaining other three dummies and their interaction with derivatives use in the model. The estimation of this model yields an interesting result. The effect of derivatives use on firm value is asymmetric. First of all, the use of derivatives has a significant positive impact on firm value when a firm is diversified. This is in line with the evidence in Lin et al. (2007) who find that diversified firms benefit from financial risk management. Moreover, this effect is not affected by the presence or absence of family owners, but is a bit stronger among diversified non-family firms with an estimated coefficient of 0.3724, which is significant at the 1% level.

Interestingly, as demonstrated by the coefficient of Derivatives Use (with no interaction), undiversified family firms (the base group) do not experience any significant value change when they use derivatives. However, the use of financial derivatives adds value to family firms that are diversified and, therefore, have the ability to construct operational hedges. This finding is similar to evidence in past studies by Lim and Wang (2007) and Lin et al. (2007) who determine that diversified firms benefit from financial risk management as derivative usage lowers information asymmetry thereby reducing the negative valuation effects of diversification.

VII. Robustness Tests

A. Alternative Estimation Methodologies

In this subsection, we present the results of several robustness checks aimed at ensuring that the previous findings are not due to the particular estimation methodology used. According to Petersen (2009), the most common methods used in recently published finance papers are the Fama and MacBeth (1973) procedure, fixed-effect regressions, and models with cluster-adjusted standard errors. Petersen (2009) goes on to confirm that in many cases, any chosen method may be incorrect and yield different results. Therefore, we examine whether our evidence holds by using all of the aforementioned methods. In addition, Campa and Kedia (2002), in their quest for the true relationship between diversification and firm value, also test other models, such as panel regressions and a self-selection model. Accordingly, our robustness tests include a fixed-effects model (Column [1]), a random-effects model (Column [2]), a self-selection model (Column [3]), and a model where statistical significance is computed using standard errors clustered at the firm level and adjusted for heteroskedasticity using White's (1980) procedure (Column [4]). Finally, we estimate a model using only the first year observation of each firm (Column [5]) to assess whether or not the previous results are driven by the existence of multiple observations on the same firms. (11) The results of these robustness checks are reported in Table IX. Panel A contains the operational hedging (diversification) models, whereas Panel B provides the financial hedging (derivatives use) models. We find that our results remain essentially unchanged under all of the other estimation methods. Firm excess value (Excess Value) is positively associated with family firms and hedging policy, but negatively related with the interaction term.

B. Expanded Tests with 2004 Data

As previously mentioned in Section I, our analysis of financial hedging policies is restricted to a five-year period (1992-1996) for which complete annual derivatives-use data is available by Swaps Monitor. We assume that using a particular time period does not drive our results. To verify whether this assumption is reasonably conservative, we replicate our main tests involving financial hedging policies for a later sample period.

In light of the widely documented difficulties associated with the construction of derivatives-use data sets, we collected information on our sample firms for a single additional year, 2004. (12) The choice of the year was also determined by the availability of information on family firms, a list of which we obtained from the November 10,2003 issue of Business Week. We hand-collected derivatives-use information from 10-K statements applying the same criteria as we had in our main sample. We exclude any firms from this new sample if their equity values are less than $20 million or they are regulated utilities, depository institutions, or holding, or other investment firms.

Using the new data, we replicate the univariate and multivariate tests of Tables VI and VII and present an abbreviated summary of the results in Table X. The results are qualitatively similar to ones we find using the main sample covering the earlier time period, although given the smaller sample size, they are expectedly a bit weaker in statistical terms. Overall, the results indicate that active financial hedging policy in family-owned firms is value destroying. Family owners may be influencing hedging policy to reduce their own undiversified wealth portfolio. Overall, the evidence in Table X confirms that our earlier test results hold in a later sample period and are not driven by a specific time period.

VIII. Conclusions

In recent years, the finance literature has highlighted the importance of founding family presence in corporate boards, and has provided some evidence that family firms perform better than non-family firms. However, some other studies provide evidence supporting the view that founding family presence is associated with greater agency costs and, as such, is not characteristic of an optimal ownership structure.

In this paper, we add to the literature by providing evidence to reconcile the seemingly contradictory evidence from these two streams from prior studies. Our empirical methodology involves examining the market value impact of family firms' risk management policies, which are expected to be reflective of agency conflicts that may exist within family firms. Our analysis utilizes the Berger and Ofek (1995) excess value measure and accounts for both dimensions of corporate risk management policies: 1) operational hedging (proxied here by the degree of corporate diversification) and 2) financial hedging (proxied by the extent and type of derivatives' use).

We determine that in the absence of an active risk management policy, family firms exhibit higher values than non-family firms, supporting the argument that founding families lower agency costs by monitoring managers more efficiently and by having longer investment horizons. We also find that risk management policies are value increasing across both operational and financial dimensions. However, our evidence indicates that in the presence of founding families, risk management policies do not add any value to the firms.

Collectively, our results suggest that founding families' presence is beneficial for firms as it can improve monitoring and lead to greater investment efficiency. However, family firms may also suffer from greater agency problems that are reflected in inefficient hedging policies. Our empirical results shed new light on the two faces of founding family presence in firms implying that the impact of family ownership on firm value should be examined more carefully.

References

Aabo, T., M.A. Hansen, and C. Pantzalis, 2012, "Corporate Foreign Exchange Speculation and Integrated Risk Management," Managerial Finance 38, 729-751.

Aabo, T. and C. Pantzalis, 2011, "In or Out: The Effect of Euro Membership on the Exercise of Real Business Options," European Journal of Finance 17, 259-284.

Allayannis, G., J. Ihrig, and J.P. Weston, 2001, "Exchange-Rate Hedging: Financial vs. Operational Strategies," American Economic Review 91, 391-395.

Allayannis, G. and M.J. Schill, 2010, "Financial Policies and Hedging," University of Virginia Working Paper.

Allayannis, G. and J.P. Weston, 2001, "The Use of Foreign Currency Derivatives and Firm Market Value," Review of Financial Studies 14, 243-276.

Amihud, Y. and B. Lev, 1981, "Risk Reduction as a Managerial Motive for Conglomerate Mergers," Bell Journal of Economics 12, 605-617.

Anderson, R.C., A. Duru, and D.M. Reeb, 2009, "Founders, Heirs, and Corporate Opacity in the US," Journal of Financial Economics 92, 205-222.

Anderson, R.C. and D.M. Reeb, 2003a, "Founding Family Ownership and Firm Performance: Evidence from the S&P 500f Journal of Finance 58, 1301-1328.

Anderson, R.C. and D.M. Reeb, 2003b, "Founding-Family Ownership, Corporate Diversification, and Firm Leverage," Journal of Law and Economics 46, 653-684.

Anderson, R.C. and D.M. Reeb, 2004, "Board Composition: Balancing Family Influence in S&P 500 Firms," Administrative Science Quarterly 49, 207-237.

Antia, M., C. Pantzalis, and J.C. Park, 2010, "CEO Decision Horizon and Firm Performance: An Empirical Investigation," Journal of Corporate Finance 16, 288-301.

Bartram, S.M., 2004, "Linear and Nonlinear Foreign Exchange Rate Exposures of German Nonfinancial Corporations," Journal of International Money and Finance 23, 673-689.

Berger, P.G. and E. Ofek, 1995, "Diversification's Effect on Firm Value," Journal of Financial Economics 37, 39-65.

Bodnar, G.M., G.S. Hayt, and R.C. Marston, 1996, "1995 Wharton Survey of Derivatives Usage by US Non-Financial Firms," Financial Management 25, 113-133.

Bodnar, G.M., G.S. Hayt, and R.C. Marston, 1998, "1998 Wharton Survey of Derivatives Usage by US Non-Financial Firms," Financial Management 27, 70-91.

Bradley, M., A. Desai, and E.H. Kim, 1988, "Synergistic Gains from Corporate Acquisitions and Their Division between the Stockholders and Target and Acquiring Firms," Journal of Financial Economics 21, 3-40.

Breeden, D. and S. Viswanathan, 1998, "Why Do Firms Hedge? An Asymmetric Information Model," Duke University Working Paper.

Campa, J.M. and S. Kedia, 2002, "Explaining the Diversification Discount," Journal of Finance 57, 1731-1762.

Carter, D., C. Pantzalis, and B. J. Simkins, 2001, "Firmwide Risk Management of Foreign Exchange Exposure by US Multinational Corporations," Oklahoma State University Working Paper.

Carter, D., D.A. Rogers, and B.J. Simkins, 2006, "Does Fuel Hedging Make Economic Sense? The Case of the US Airline Industry," Financial Management 35, 53-86.

DaDalt, P., G.D. Gay, and J. Nam, 2002, "Asymmetric Information and Corporate Derivatives Use," Journal of Futures Markets 22, 241-267.

DeMarzo, P. and D. Duffie, 1995, "Corporate Incentives for Hedging and Hedge Accounting," Review of Financial Studies 8, 743-771.

Demsetz, H., 1983, "The Structure of Ownership and the Theory of the Firm," Journal of Law and Economics 25, 375-390.

Demsetz, H. and K. Lehn, 1985, "The Structure of Corporate Ownership: Causes and Consequences," Journal of Political Economy 93, 1155-1177.

Dolde, W., 1993, "The Trajectory of Corporate Financial Risk Management," Journal of Applied Corporate Finance 6, 33-41.

Dolde, W., 1995, "Hedging, Leverage, and Primitive Risk," Journal of Financial Engineering A, 187-216.

Dolde, W. and D.R. Mishra, 2007, "Firm Complexity and Foreign Exchange Derivatives Use," Quarterly Journal of Business and Economics 46, 3-22.

Doukas, J., C. Kim, and C. Pantzalis, 2005, "The Two Faces of Analyst Coverage," Financial Management 34, 99-126.

Fama, E. and M. Jensen, 1983, "Separation of Ownership and Control," Journal of Law and Economics 26, 301-325.

Fama, E.F. and J.D. MacBeth, 1973, "Risk, Return and Equilibrium: Empirical Tests ''Journal of Political Economy 81, 607-636.

Frankel, R. and C.M.C. Lee, 1998, "Accounting Valuation, Market Expectation and Cross-Sectional Stock Returns," Journal of Accounting and Economics 25, 283-319.

Froot, K., D. Scharfstein, and J. Stein, 1993, "Risk Management: Coordinating Corporate Investment and Financing Policies," Journal of Finance 48, 1624-1658.

Geczy, C., B. A. Minton, and C. Schrand, 1997, "Why Firms Use Currency Derivatives," Journal of Finance 52, 1323-1354.

Geczy, C., B.A. Minton, and C. Schrand, 2007, "Taking a View: Corporate Speculation, Governance, and Compensation," Journal of Finance 62, 2405-2443.

Graham, J.R., M.L. Lemmon, and J.G. Wolf, 2002, "Does Corporate Diversification Destroy Value?" Journal of Finance 57, 695-720.

Graham, J.R. and D.A. Rogers, 2002, "Do Firms Hedge in Response to Tax Incentives?" Journal of Finance 57, 815-839.

James, H., 1999, "Owner as Manager, Extended Horizons and the Family Firm," International Journal of the Economics of Business 6, 41-56.

Jensen, M.C., 1986, "Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers," American Economic Review 76, 323-330.

Jensen, M.C. and K.J. Murphy, 1990, "Performance Pay and Top-Management Incentives," Journal of Political Economy 98, 225-264.

Lamont, O., 1997, "Cash Flow and Investment: Evidence from Internal Capital Markets," Journal of Finance 52, 83-110.

Lang, L.H.P. and R.M. Stulz, 1994, "Tobin's Q, Corporate Diversification, and Firm Performance," Journal of Political Economy 102, 1248-1280.

Lim, S.S. and H. Wang, 2007, "The Effect of Financial Hedging on the Incentives for Corporate Diversification: The Role of Stakeholder Firm-Specific Investments," Journal of Economic Behavior and Organization 62, 640-656.

Lin, B.J., C. Pantzalis, and J.C. Park, 2007, "Corporate Use of Derivatives and Excess Value of Diversification," Journal of Banking and Finance 31, 889-913.

Lin, B.J., C. Pantzalis, and J.C. Park, 2009, "Derivatives Use, Information Asymmetry and MNC Post-Acquisition Performance," Financial Management 38, 631-661.

Lin, B.J., C. Pantzalis, and J.C. Park, 2010, "Corporate Hedging Policy and Equity Mispricing," Financial Review 45, 803-824.

Lins, K. and H. Servaes, 1999, "International Evidence on the Value of Corporate Diversification," Journal of Finance 54, 2215-2240.

Loughran, T. and A.M. Vijh, 1997, "Do Long-Term Shareholders Benefit from Corporate Acquisitions?" Journal of Finance 52, 1765-1790.

Mansi, S. A. and D.M. Reeb, 2002, "Corporate Diversification: What Gets Discounted?" Journal of Finance 57, 2167-2183.

Mian, S.L., 1996, "Evidence on Corporate Hedging Policy," Journal of Financial and Quantitative Analysis 31, 419-439.

Mork, R., A. Shleifer, and R.W. Vishny, 1990, "Do Managerial Objectives Drive Bad Acquisitions?" Journal of Finance 45, 31-48.

Nance, D.R., C.W. Smith, and C.W. Smithson, 1993, "On the Determinants of Corporate Hedging," Journal of Finance 48, 267-285.

Ohlson, J.A., 1995, "Earnings, Book Values, and Dividends in Security Valuation," Contemporary Accounting Research 11, 661-687.

Pantzalis, C., B.J. Simkins, and P.A. Laux, 2001, "Operational Hedges and the Foreign Exchange Exposure of US Multinational Corporations," Journal of International Business Studies 32, 793-812.

Petersen, M.A., 2009, "Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches," Review of Financial Studies 22, 435-480.

Rajan, R., H. Servaes, and L. Zingales, 2000, "The Cost of Diversity: The Diversification Discount and Inefficient Investment," Journal of Finance 55, 35-80.

Rau, R. and T. Vermaelen, 1998, "Glamour, Value and the Post-Acquisition Performance of Acquiring Firms," Journal of Financial Economics 49, 223-253.

Scharfstein, D.S. and J.C. Stein, 2000, "The Dark Side of Internal Capital Markets: Divisional Rent-Seeking and Inefficient Investment," Journal of Finance 55, 2537-2564.

Servaes, H., 1996, "The Value of Diversification during the Conglomerate Merger Wave," Journal of Finance 51, 1201-1225.

Shapiro, A. and S. Titman, 1985, "An Integrated Approach to Corporate Risk Management," Midland Corporate Finance Journal 3, 41-56.

Shin, H.-H. and R.M. Stulz, 1998, "Are Internal Capital Markets Efficient?" Quarterly Journal of Economics 113,531-552.

Shleifer, A. and R.W. Vishny, 1986, "Large Shareholders and Corporate Control," Journal of Political Economy 94, 461 -489.

Shleifer, A. and R.W. Vishny, 1997, "A Survey of Corporate Governance," Journal of Finance 52, 737-783.

Singh, M., W.N. Davidson, and J.-A. Suchard, 2003, "Corporate Diversification Strategies and Capital Structure," Quarterly Review of Economics and Finance 43, 147-167.

Smith, C. and R.M. Stulz, 1985, "The Determinants of Firms' Hedging Policies," Journal of Financial and Quantitative Analysis 20, 391-405.

Stein, J., 1988, "Takeover Threats and Managerial Myopia," Journal of Political Economy 96, 61-80.

Stein, J., 1989, "Efficient Capital Markets, Inefficient Firms: A Model of Myopic Corporate Behavior," Quarterly Journal of Economics 103, 655-669.

Stulz, R.M., 1984, "Optimal Hedging Policies," Journal of Financial and Quantitative Analysis 19,127-140.

Stulz, R.M., 1990, "Managerial Discretion and Optimal Financing Policies," Journal of Financial Economics 26, 3-27.

Tufano, P., 1996, "Who Manages Risk? An Empirical Examination of Risk Management Practices in the Gold Mining Industry," Journal of Finance 51, 1097-1137.

Villalonga, B., 2004a, "Does Diversification Cause the 'Diversification Discount'?" Financial Management 33, 5-27.

Villalonga, B., 2004b, "Diversification Discount of Premium? New Evidence from the Business Information Tracking Series," Journal of Finance 59, 479-506.

Villalonga, B. and R. Amit, 2006, "How Do Family Ownership, Control, and Management Affect Firm Value?" Journal of Financial Economics 80, 385-417.

Wernerfelt, B. and C.A. Montgomery, 1988, "Tobin's Q and the Importance of Focus in Firm Performance," American Economic Review 78, 246-250.

White, H., 1980, "A Heteroskedasticity Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity," Econometrica 48, 817-838.

Whited, T.M., 2001, "Is it Inefficient Investment that Causes the Diversification Discount?" Journal of Finance 56, 1667-1992.

(1) Firms that are managed or controlled by their founders or by the founders' families and heirs are, hereafter, referred to as family firms.

(2) Many theoretical studies (Stulz, 1984; Shapiro and Titman, 1985; Smith and Stulz, 1985; Froot, Scharfstein, and Stein, 1993; DeMarzo and Duffie, 1995) have demonstrated that in the presence of market imperfections, such as taxes, financial distress costs, and agency conflicts, corporate hedging policy becomes relevant (i.e., hedging decisions may have a positive impact on firm valuation).

(3) There are at least two possible reasons advanced by the diversification discount literature for the poor performance of diversified firms. First, diversified firms may suffer from a capital misallocation problem (Lamont, 1997; Shin and Stulz, 1998; Scharfstein and Stein, 2000; Rajan, Servaes, and Zingales, 2000). Second, in addition, managers pursuing their own interests have incentives to diversify their firms by undertaking nonvalue maximizing acquisitions (Amihud and Lev, 1981; Mork, Shleifer, and Vishny, 1990). Other studies propose alternative reasons for managers' preference for larger firm sizes, such as increasing their power and prestige (Jensen, 1986; Stulz, 1990) or increasing their compensation (Jensen and Murphy, 1990). Furthermore, the takeover and asset sell-off literature provides additional evidence to support the value reducing effects of diversification (Bradley, Desai, and Kim, 1988; Berger and Ofek, 1995; Loughran and Vijh, 1997; Rau and Vermaelen, 1998).

(4) This view is also supported by the empirical results of Lin, Pantzalis, and Park (2009, 2010). Lin et al. (2010) find that firms' use of derivatives is negatively associated with stock mispricing. Lin et al. (2009) examine the impact of derivatives use on the post-M&A (mergers & acquisitions) performance of US MNCs and find that acquirers with derivative policies that are more comprehensive and sophisticated outperform those with less comprehensive and sophisticated policies. They, in turn, outperform acquirers with no existing policies in place. The use of derivatives can lower information asymmetry and related agency problems.

(5) We choose the year 2004 for our extended study on hedging policies as we are able to obtain the list of family firms from the November 10, 2003 issue of BusinessWeek. As we do for the main sample, we use S&P 500 firms and exclude any firms from this new sample if their equity values are less than $20 million or they are regulated utilities, depository institutions, and holding, or other investment firms. We hand-collected derivatives-use information from 10-K statements.

(6) We also compute excess value based on the four digit SIC code industry classification and find that the results are similar to those when we use Fama-French 48 industry classifications. These results are omitted for the sake of brevity, but are available from the authors upon request.

(7) Campa and Kedia (2002) extend the set of control variables beyond those originally used in the BO model to account for endogeneity in firms' decision to diversify.

(8) In addition, following Campa and Kedia (2002), we also conduct panel regressions and a self-selection model. We report these results in Table IX.

(9) For a justification of the use of leverage as a control variable and its effect on diversification, see Singh, Davidson, and Suchard (2003).

(10) The sample size in the 2SLS tests is reduced to 1,450 observations due to missing values of foreign sales and historical average values of some of the variables in the probit estimation performed in the first-stage regression. We obtain foreign sales data for only 1,861 firm-years as reported in Table I. Also, historical variables that require preceding five- year information caused a reduction in the sample size.

(11) We do not use the time-series average of cross-sectional annual regressions as introduced in Fama and MacBeth (1973) since our data period (1992-1999) is too short to have meaningful statistical interpretations.

(12) Collecting information on off-balance sheet instruments is a considerably time-consuming task. As such, most studies in the literature rely on surveys of a small population of firms (Nance, Smith, and Smithson, 1993; Dolde, 1993, 1995; Bodnar, Hayt, and Marston, 1996, 1998). Other studies have been able to examine firms' financial reports (Mian, 1996; Tufano, 1996; Dolde and Mishra, 2007), but their sample sizes are usually limited to either a single year and/or fewer firms. Therefore, the sample sizes from these previous papers are quite small. For example, Tufano (1996) collects derivatives information for 48 firms in the mining industry, while Nance et al. (1993) have 169 firms in their final sample.

We are especially grateful to Raghavendra Rau (Editor) and an anonymous reviewer for their many insightful and constructive suggestions.

Chansog Kim, Christos Pantzalis, and Jung Chul Park *

* Chansog (Francis) Kim is an Associate Professor of Accounting at Wayne State University in Detroit, MI. Christos Pantzalis is a Professor of Finance and the Bank of America Professor at the University of South Florida in Tampa, FL. Jung Chul Park is an Associate Professor of Finance and the McLain Family Professor at Auburn University in Auburn, AL.

Table I. Descriptive Statistics Panel A reports descriptive statistics for our sample firms. Family is an indicator that takes a value of one if the founding family is present and zero otherwise. Excess value is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to sales. Excess Value (TA) is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to total assets. Excess Value (RI) = the measure of firm's excess value based on Ohlson's (1995) residual income. Excess Value (MB) is the measure of firm's excess value based on the industry-adjusted market-to-book ratio. Diversification is an indicator that takes a value of one if the firm reports two or more business segments and zero if it reports only one segment. Number of Segments is the number of business segments the firm reports. Derivatives Use is a dummy variable that takes a value one if a firm uses any kind of derivatives against foreign exchange risk or interest rate risk, and otherwise zero. Number of Types is the number of types of derivatives used (i.e., FX and-or IR), or none and takes values from zero to two. Number of Contracts is the number of different derivative contracts used by the firm and based on the Swaps Monitor's database. It can take values from zero to seven. Contracted Dollar Amount is the total dollar amount of derivatives used. Size is the log of total assets. Leverage is the ratio of long-term debt to total assets. Foreign Sales is an indicator that takes a value of one if the firm is incorporated in foreign sales and zero otherwise. Panel B provides detailed descriptive statistics of sample firms' policy of derivatives use. The Swaps Monitor indicates the usage of derivatives, as well as the notional amounts for seven different contracts spanning two general types of derivatives: 1) interest rate (IR) derivatives and 2) foreign exchange (FX) derivatives. The seven different contracts are: 1) IR-options, 2) IR-swaps, 3) IR-forwards-futures, 4) FX-options, 5) FX-swaps, 6) FX-futures, and 7) FX-forwards. N Mean SD Minimum Panel A. Descriptive Statistics of the Variables Examined Family presence Family 2,539 0.3324 0.4712 0.0000 Firm valuation Excess Value 2,539 0.0179 0.6497 -2.2334 Excess Value (TA) 2,468 0.1074 0.5248 -1.9591 Excess Value (RJ) 2,194 0.6576 0.5835 -1.8192 Excess Value (MB) 2,538 0.0389 0.3851 -1.4189 Operational hedging Diversification 2,498 0.6105 0.4877 0.0000 Number of Segments 2,498 2.3415 1.4571 1.0000 Financial hedging Derivatives Use 958 0.7641 0.4248 0.0000 Number of Types 958 1.1054 0.7528 0.0000 Number of Contracts 958 1.6733 1.3824 0.0000 Contracted Dollar Amount 765 878.9923 2288.7960 0.0000 Firm characteristics Size 2,539 22.2309 1.3500 16.9769 EBIT/Sales 2,485 0.1078 0.0948 -0.4976 Capital Expenditure/Sales 2,501 0.0715 0.0823 0.0000 Leverage 2,538 0.1724 0.1439 0.0000 Foreign Sales 1,861 0.6862 0.4642 0.0000 25th Percentile Median Panel A. Descriptive Statistics of the Variables Examined Family presence Family 0.0000 0.0000 Firm valuation Excess Value -0.3630 0.0438 Excess Value (TA) -0.2150 0.0596 Excess Value (RJ) 0.3662 0.6776 Excess Value (MB) -0.1740 0.0000 Operational hedging Diversification 0.0000 1.0000 Number of Segments 1.0000 2.0000 Financial hedging Derivatives Use 1.0000 1.0000 Number of Types 1.0000 1.0000 Number of Contracts 1.0000 2.0000 Contracted Dollar Amount 0.0000 115.0000 Firm characteristics Size 21.3642 22.1544 EBIT/Sales 0.0586 0.0997 Capital Expenditure/Sales 0.0320 0.0525 Leverage 0.0612 0.1411 Foreign Sales 0.0000 1.0000 75th Percentile Maximum Panel A. Descriptive Statistics of the Variables Examined Family presence Family 1.0000 1.0000 Firm valuation Excess Value 0.4415 2.4340 Excess Value (TA) 0.4196 3.2696 Excess Value (RJ) 0.9471 5.7628 Excess Value (MB) 0.2031 2.7880 Operational hedging Diversification 1.0000 1.0000 Number of Segments 3.0000 10.0000 Financial hedging Derivatives Use 1.0000 1.0000 Number of Types 2.0000 2.0000 Number of Contracts 3.0000 6.0000 Contracted Dollar Amount 674.7000 31031.5000 Firm characteristics Size 23.1008 26.7277 EBIT/Sales 0.1492 0.6066 Capital Expenditure/Sales 0.0834 2.0328 Leverage 0.2567 0.8102 Foreign Sales 1.0000 1.0000 N Mean SD Median Panel B: Detailed Descriptive Statistics of Financial Hedging Use of derivatives FX 958 0.6388 0.4806 1.0000 FX-Swaps 958 0.2453 0.4305 0.0000 FX-Options 958 0.2432 0.4292 0.0000 FX-Forwards 958 0.6054 0.4890 1.0000 FX-Futures 958 0.0000 0.0000 0.0000 IR 958 0.4666 0.4991 0.0000 IR-Swaps 958 0.4113 0.4923 0.0000 IR-Options 958 0.1399 0.3470 0.0000 IR-Forwards/Futures 958 0.0282 0.1656 0.0000 All (FX and/or IR) 958 0.7641 0.4248 1.0000 Notional dollar amount of derivative contracts (million dollars) FX 779 573.5937 1674.7280 24.8000 FX-Swaps 914 166.6863 1299.1500 0.0000 FX-Options 853 116.4321 712.0757 0.0000 FX Forwards 804 397.4027 1205.9730 10.5500 FX-Futures 958 0.0000 0.0000 0.0000 IR 931 469.2555 2470.2920 0.0000 IR-Swaps 950 368.8936 1897.0810 0.0000 IR-Options 941 115.5485 1316.428 0.0000 IR-Forwards/Futures 948 55.7948 1066.6730 0.0000 All (FX and/or IR) 765 878.9923 2288.7960 115.0000 Ratio of amount of derivative contracts to total capital FX 779 0.0483 0.1674 0.0070 FX-Swaps 914 0.0065 0.0223 0.0000 FX-Options 853 0.0095 0.1126 0.0000 FX-Forwards 804 0.0336 0.0725 0.0027 FX-Futures 958 0.0000 0.0000 0.0000 IR 931 0.0434 0.1239 0.0000 IR-Swaps 950 0.0338 0.0856 0.0000 IR-Options 941 0.0081 0.0454 0.0000 IR-Forwards/Futures 948 0.0012 0.0223 0.0000 All (FX and/or IR) 765 0.0883 0.2005 0.0295 75th Percentile Maximum Panel B: Detailed Descriptive Statistics of Financial Hedging Use of derivatives FX 1.0000 1.0000 FX-Swaps 0.0000 1.0000 FX-Options 0.0000 1.0000 FX-Forwards 1.0000 1.0000 FX-Futures 0.0000 0.0000 IR 1.0000 1.0000 IR-Swaps 1.0000 1.0000 IR-Options 0.0000 1.0000 IR-Forwards/Futures 0.0000 1.0000 All (FX and/or IR) 1.0000 1.0000 Notional dollar amount of derivative contracts (million dollars) FX 321.0000 15435.0000 FX-Swaps 0.0000 25902.0000 FX-Options 0.0000 10100.0000 FX Forwards 253.2000 14818.0000 FX-Futures 0.0000 0.0000 IR 222.0000 56149.0000 IR-Swaps 200.0000 41134.0000 IR-Options 0.0000 31242.0000 IR-Forwards/Futures 0.0000 30506.0000 All (FX and/or IR) 674.7000 31031.5000 Ratio of amount of derivative contracts to total capital FX 0.0497 3.3860 FX-Swaps 0.0000 0.2766 FX-Options 0.0000 2.6263 FX-Forwards 0.0338 0.7597 FX-Futures 0.0000 0.0000 IR 0.0396 1.9398 IR-Swaps 0.0325 1.1046 IR-Options 0.0000 0.8352 IR-Forwards/Futures 0.0000 0.6690 All (FX and/or IR) 0.1037 3.4049 Table II. Characteristics of Family Firms Reported are the mean values of variables for family firms and non-family firms. A firm is classified into the group of family firms if founding family is present and into the group of non-family firms otherwise. Also reported are mean differences between family and non-family firms and corresponding t-statistics. The variables examined are defined as follows. Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to sales. Excess Value (TA) is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to total assets. Excess Value (RI) is the measure of a firm's excess value based on Ohlson's (1995) residual income. Excess Value (MB) is the measure of a firm's excess value based on the industry-adjusted market-to-book ratio. Diversification is an indicator that takes a value of one if the firm reports two or more business segments and zero if it reports only one segment. Number of Segments is the number of business segments the firm reports. Derivatives Use is a dummy variable that takes the value one if a firm uses any kind of derivatives against foreign exchange risk or interest rate risk, and otherwise zero. Number of Types is the number of types of derivatives used (i.e., FX and-or IR), or none and takes values from zero to two. Number of Contracts is the number of different derivative contracts used by the firm and based on the Swaps Monitor's database. It can take values from zero to seven. Contracted Dollar Amount is the total dollar amount of derivatives used. Size is the log of total assets. Leverage is the ratio of long- term debt to total assets. Foreign sales is an indicator that takes a value of one if the firm is incorporated in foreign sales and zero otherwise. Refer to Table I for detailed variable definitions. Family Non-family All Firms Firms Firms [N = 844] [N= 1,695] [N = 2,593] Firm valuation Excess Value 0.1505 -0.0481 0.0179 Excess Value (TA) 0.1750 0.0743 0.1074 Excess Value (RI) 0.7300 0.6207 0.6576 Excess Value (MB) 0.0734 0.0218 0.0389 Diversification (operational hedging policy) Diversification 0.4964 0.6671 0.6105 Number of Segments 1.9662 2.5275 2.3415 Financial hedging policy Derivatives Use 0.7347 0.7771 0.7641 Number of Types 0.9762 1.1627 1.1054 Number of Contracts 1.2959 1.8404 1.6733 Contracted Dollar Amount 566.2653 1026.3350 878.9923 Firm characteristics Size 21.8483 22.4215 22.2309 EBIT/Sales 0.1120 0.1057 0.1078 Capital Expenditure/Sales 0.0670 0.0737 0.0715 Leverage 0.1469 0.1851 0.1724 Foreign Sales 0.6299 0.7159 0.6862 Family- t- Non-family Statistics Firm valuation Excess Value 0.1985 *** 7.33 Excess Value (TA) 0.1007 *** 4.50 Excess Value (RI) 0.1093 *** 4.17 Excess Value (MB) 0.0516 *** 3.19 Diversification (operational hedging policy) Diversification -0.1707 *** -8.35 Number of Segments -0.5614 *** -9.22 Financial hedging policy Derivatives Use -0.0424 -1.43 Number of Types -0.1865 *** -3.56 Number of Contracts -0.5444 *** -5.71 Contracted Dollar Amount -460.0695 *** -2.60 Firm characteristics Size -0.5731 *** -10.28 EBIT/Sales 0.0062 1.54 Capital Expenditure/Sales -0.0067 * -1.93 Leverage -0.0382 *** -6.35 Foreign Sales -0.0861 ** -3.82 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table III. Propensity to Diversify and to Hedge This table reports the estimated coefficients and marginal effects for a probit model using the diversification and derivatives-use indicators as dependent variables. The variables are defined as follows. Diversification is an indicator that takes a value of one if the firm reports two or more business segments and zero if it reports only one segment. Derivatives Use is a dummy variable that takes a value one if a firm uses any kind of derivatives against foreign exchange risk or interest rate risk, and otherwise zero. Family is an indicator that takes a value of one if founding family is present and zero otherwise. Size is the log of total assets. Fraction of Conglomerates is the fraction of c in the industry that is a conglomerate. Fraction of Conglomerates ' Sales is the fraction of industry sales accounted for by conglomerates. Number of M&As is the number of announced mergers and acquisitions in the year. Volume of M&As is the US dollar value of announced mergers and acquisitions in the year. GDP Growth is the growth rate in real GDP. The historical average of the variable is computed as the average of the preceding five years. NYSE is an indicator that takes a value of one if the firm is exchanged in the New York Stock Exchange. Foreign Sales is an indicator that takes one if the firm is incorporated in foreign sales and zero otherwise. Z-statistics are shown in parentheses. Dependent Variable: Diversification Coeff. Z- Marginal Statistics Effect ([DELTA]D/ [DELTA]X) Family -0.3059 *** -3.65 -0.1160 Size 0.1548 0.65 0.0580 EBIT/Sales -0.5564 -0.80 -0.2086 Capital Expenditure/Sales 0.8480 0.71 0.3179 Size (1 lag) 0.0428 0.12 0.0161 EBIT/Sales (1 lag) -0.5643 -0.74 -0.2116 Capital Expenditure/Sales -1.3242 -0.86 -0.4965 (1 lag) Size (2 lag) -0.0054 -0.02 -0.0020 EBIT/Sales (2 lag) -1.9737 -2.67 -0.7400 Capital Expenditure/Sales 1.3558 1.15 0.5083 (2 lag) Fraction of Conglomerates 1.6652 ** 3.48 0.6243 Fraction of Conglomerates' 1.5485 *** 5.78 0.5806 Sales Number of M&As 0.0004 0.22 0.0001 Volume of M&As -0.0000 -0.28 -0.0000 GDP Growth -0.5009 -0.12 -0.1878 GDP Growth (1 lag) -1.0964 -0.29 -0.4110 Size (historical average) 0.4836 *** 3.13 0.1813 EBIT/Sales (historical -0.2381 ** -2.16 -0.0893 average) Capital Expenditure/Sales -0.3138 *** -3.68 -0.1177 {historical average) NYSE 0.8729 *** 4.63 0.3373 Foreign Sales 0.1229 1.45 0.0464 Intercept -6.6306 *** -3.65 N 1,451 Pseudo [R.sup.2] 0.2435 Dependent Variable: Derivatives Use Coeff. Z- Marginal Statistics Effect ([DELTA]D/ [DELTA]X) Family -0.0459 -0.34 -0.0125 Size -0.8873 ** -2.00 -0.2391 EBIT/Sales 0.8577 0.59 0.2311 Capital Expenditure/Sales -2.0623 -0.93 -0.5556 Size (1 lag) 1.3677 ** 2.08 0.3685 EBIT/Sales (1 lag) -0.3365 -0.21 -0.0907 Capital Expenditure/Sales 1.1311 0.41 0.3047 (1 lag) Size (2 lag) -0.2764 -0.59 -0.0745 EBIT/Sales (2 lag) 0.9204 0.65 0.2480 Capital Expenditure/Sales 1.0499 0.51 0.2829 (2 lag) Fraction of Conglomerates -1.5901 ** -1.98 -0.4284 Fraction of Conglomerates' 1.1263 ** 2.48 0.3035 Sales Number of M&As 0.0015 0.60 0.0004 Volume of M&As -0.0000 ** -2.28 -0.0000 GDP Growth -6.1644 -1.05 -1.6608 GDP Growth (1 lag) 10.5103 0.88 2.8317 Size (historical average) 0.2074 0.64 0.0559 EBIT/Sales (historical 0.1267 0.63 0.0341 average) Capital Expenditure/Sales -0.2416 -1.24 -0.0651 {historical average) NYSE 0.1225 0.49 0.0346 Foreign Sales 0.7576 *** 5.91 0.2337 Intercept -5.0037 -1.40 N 671 Pseudo [R.sup.2] 0.2022 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table IV. Family Firms, Diversification, and Valuation Panel A compares firms' excess value of subgroups classified based on family presence (Family) and industrial diversification (Diversification). Also reported are mean differences between subgroups and corresponding t-statistics (in parentheses). Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to sales. Family is an indicator that takes a value of one if founding family is present and zero otherwise. Diversification is an indicator that takes a value of one if the firm reports two or more business segments and zero if it reports only one segment. Panel B provides the cross-sectional regressions of excess value on the family-firm indicator, the diversification indicator, and the other firm characteristics. Size is the log of total assets. Note that, t-statistics are shown in parentheses. Panel A. Comparison of Excess Value Family Non-family Firms Firms (Family = 1) (Family = 0) Diversified Firms 0.0900 -0.0720 (Diversification = 1) [N =411] [N= 1,114] Nondiversified Firms 0.2001 -0.0049 (Diversification = 0) [N = 417] [N 556] All Firms 0.1455 -0.0496 [N = 828] [N= 1,670] Diversified- -0.1101 *** -0.0671 * Nondiversified (-2.62) (-1.95) Family- All Firms Non-family Diversified Firms -0.0283 0.1620 *** (Diversification = 1) [N = 1,525] (4.40) Nondiversified Firms 0.0830 0.2050 *** (Diversification = 0) (4.84) All Firms 0.0150 0.1951 *** [N = 2,498] (7.12) Diversified- -0.1113 *** Nondiversified (4.18) Panel B. Regression Analysis Dependent Variable: [1] [2] Excess Value Family 0.1942 *** (7.38) Diversification -0.0743 *** (-2.84) Family * Diversification Size 0.0017 -0.0058 (0.18) (-0.59) EBIT/Sales 2.4964 *** 2.5502 *** (19.30) (19.35) Capital Expenditure/Sales 0.5913 *** 0.4932 ** (4.06) (3.24) Intercept -0.3977 * -0.1235 (-1.91) (-0.59) N 2,459 2,429 Adj. [R.sup.2] 0.1615 0.1467 Dependent Variable: [3] [4] Excess Value Family 0.1850 ** 0.2431 *** (6.95) (6.10) Diversification -0.0539 ** -0.0169 (-2.07) (-0.52) Family * Diversification -0.1035 * (-1.96) Size 0.0064 0.0067 (0.65) (0.68) EBIT/Sales 2.4891 *** 2.4932 *** (19.02) (19.06) Capital Expenditure/Sales 0.5343 *** 0.5654 *** (3.54) (3.73) Intercept -0.4622 ** -0.4962 ** (-2.17) (-2.32) N 2,429 2,429 Adj. [R.sup.2] 0.1630 0.1640 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table V. Family Firms, Diversification, and Valuation: Two-stage Least-Squares Regressions This table reports the 2SLS regressions of excess value on the family-firm indicator, the diversification indicator, and the other firm characteristics. Based on Campa and Kedia (2002), the first step involves a probit model wherein the diversification dummy is estimated, and we report the second-stage regression result in this table. The variables are defined as follows. Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to sales. Columns [5] to [7] use alternative measures of excess value as a dependent variable: Excess Value (TA) is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to total assets instead of sales; Excess Value (RI) is the measure of a firm's excess value based on Ohlson's (1995) residual income; and Excess Value (MB) is the measure of a firm's excess value based on the industry-adjusted market-to-book ratio. Family is an indicator that takes a value of one if founding family is present and zero otherwise. Diversification is an indicator that takes a value of one if the firm reports two or more business segments and zero if it reports only one segment. Size is the log of total assets. Leverage is the ratio of long-term debt to total assets. Note that, t-statistics are shown in parentheses. Dependent Excess Value Variable: [1] [2] Family 0.1608 ** (6.17) Diversification 0.2220 *** (3.19) Family * Diversification Size 0.4483 ** 0.4234 (2.08) (1.51) EBIT/Sales 1.3227 *** 1.5120 *** (7.48) (5.76) Capital 0.0882 -0.1643 Expenditure/ (0.41) (-0.33) Sales Size (1 lag) -0.2196 ** -0.2727 * (-2.05) (-1.90) EBIT/Sales 0.5541 *** 0.3365 (1 lag) (3.04) (1.13) Capital 0.1949 0.4837 Expenditure/ (0.61) (0.75) Sales (1 lag) Size (2 lag) -0.0549 -0.1844 * (-0.75) (-1.86) EBIT/Sales 0.9240 ** 1.2941 *** (2 lag) (5.38) (4.68) Capital 0.4714 * 0.2660 Expenditure/ (1.66) (0.56) Sales (2 lag) Leverage -0.5570 *** -0.6427 *** Size (2) (-6.27) (-4.87) -0.0039 0.0002 (-0.85) (0.03) Intercept -2.2600 0.2304 (-0.98) (0.08) N 2,408 1,450 Adj. [R.sup.2] 0.2120 0.2174 Dependent Excess Value Variable: [3] [4] Family 0.1812 *** 0.5216 *** (5.43) (6.73) Diversification 0.2606 *** 0.5114 *** (3.76) (5.95) Family * -0.5880 *** Diversification (-4.86) Size 0.4127 0.5193 * (1.49) (1.88) EBIT/Sales 1.5612 *** 1.5545 *** (6.01) (6.03) Capital -0.0020 0.0480 Expenditure/ (0.00) (0.10) Sales Size (1 lag) -0.2617 * -0.2702 * (-1.84) (-1.91) EBIT/Sales 0.2806 0.3087 (1 lag) (0.95) (1.06) Capital 0.4727 0.5472 Expenditure/ (0.74) (0.86) Sales (1 lag) Size (2 lag) -0.1622 * -0.1516 (-1.65) (-1.56) EBIT/Sales 1.2763 *** 1.2823 *** (2 lag) (4.66) (4.72) Capital 0.1596 0.1215 Expenditure/ (0.34) (0.26) Sales (2 lag) Leverage -0.5925 *** -0.5621 *** Size (2) (-4.52) (-4.32) 0.0000 -0.0025 (0.00) (-0.43) Intercept -0.2653 -1.6037 (-0.09) (-0.54) N 1,450 1,450 Adj. [R.sup.2] 0.2326 0.2445 Dependent Excess Excess Excess Variable: Value (TA) Value (R1) Value (MB) [5] [6] [7] Family 0.1968 *** 0.1816 ** 0.1716 *** (3.34) (2.40) (3.97) Diversification 0.2171 *** 0.2837 *** 0.0909 * (3.36) (3.39) (1.90) Family * -0.3342 *** -0.2043 * -0.2733 *** Diversification (-3.65) (-1.73) (-4.05) Size 0.9572 *** 0.0132 1.0874 *** (4.59) (0.05) (7.06) EBIT/Sales 1.1939 *** -1.1407 *** 0.5655 *** (6.07) (-4.46) (3.93) Capital 0.0774 1.1803 ** -0.1209 Expenditure/ (0.21) (2.48) (-0.44) Sales Size (1 lag) -0.1378 -0.0102 -0.0936 (-1.30) (-0.07) (-1.19) EBIT/Sales 0.1975 -0.2659 -0.1518 (1 lag) (0.89) (-0.93) (-0.93) Capital 0.3181 -0.0922 0.4729 Expenditure/ (0.67) (-0.15) (1.33) Sales (1 lag) Size (2 lag) -0.0672 -0.1233 -0.1252 ** (-0.92) (-1.30) (-2.31) EBIT/Sales 0.3603 0.7105 *** 0.0742 (2 lag) (1.76) (2.67) (0.49) Capital 0.0005 0.3283 -0.1240 Expenditure/ (0.00) (0.72) (-0.48) Sales (2 lag) Leverage -1.6708 *** -0.6553 *** -1.2172 *** Size (2) (-17.08) (-5.11) (-16.79) -0.0180 *** 0.0024 -0.0193 *** (-4.06) (0.42) (-5.90) Intercept -7.8146 *** 2.0328 -9.6756 *** (-3.46) (0.69) (-5.81) N 1,435 1,437 1,450 Adj. [R.sup.2] 0.3400 0.0463 0.2685 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table VI. Family Firms and Financial Hedging Policy: Comparisons of Excess Value This table compares a firm's Excess Value of subgroups. Also reported are mean differences between subgroups and corresponding f-statistics (in parentheses). Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995). Family is an indicator that takes a value of one if founding family is present and zero otherwise. Derivatives Use is a dummy variable that takes a value of one if a firm uses any kind of derivatives against foreign exchange risk or interest rate risk, and otherwise zero. Number of Types is the number of types of derivatives used (i.e., FX and-or IR), or none and takes values from zero to two. Number of Contracts is the number of different derivative contracts used by the firm and based on the Swaps Monitor's database and takes values from zero to seven. Contracted Dollar Amount is the total dollar amount of derivatives used. Note that, t-statistics are shown in parentheses. [A] Family [B] Non-family Firms Firms (Family = 1) (Family = 0) Panel A. Using Derivatives Use [1] Users of derivatives 0.1491 -0.0709 (Derivatives Use = 1) [N = 216] [A = 516] [2] Nonusers of derivatives 0.1677 -0.3013 (Derivatives Use = 0) [A = 78] [A = 148] All Firms 0.1540 -0.1223 [N = 294] [A = 664] [1] to [2] -0.0186 0.2304 *** (-0.24) Panel B. Using Number of Types [1] Firms using both FX and IR 0.1333 -0.0172 (Number of Types = 2) [N = 71] [A = 256] [2] Firms using either FX or IR 0.1568 -0.1238 (Number of Types = 1) [A = 145] [A = 260] [3] Firms using no derivatives 0.1677 -0.3013 (Number of Types = 0) [A = 78] [A = 148] [1] to [3] -0.0344 0.2841 *** (-0.35) (4.16) Panel C. Using Number of Contracts [1] Firms using 4 to 6 contracts 0.2219 0.0723 (Number of Contracts = 4 to 6) [A = 13] [A = 95] [2] Firms using 1 to 3 contracts 0.1444 -0.1032 (Number of Contracts = 1 to 3) [A = 203] [A = 421] [3] Firms using no derivatives 0.1677 -0.3013 Number of Contracts = 0) [A = 78] [A = 148] [1] to [3] 0.0542 0.3736 *** (0.30) (4.11) Panel D. Using Contracted Dollar Amount [1] Firms using a larger dollar 0.1826 -0.0744 amount of derivatives [A = 50] [A = 219] [2] Firms using a smaller dollar 0.0964 -0.0518 amount of derivatives [A = 117] [A = 153] [3] Firms using no derivatives 0.1677 -0.3013 [A = 78] [A = 148] [1] to [3] -0.0149 0.2269 *** (-0.14) (3.22) All [A] and Firms [B] Panel A. Using Derivatives Use [1] Users of derivatives -0.0060 0.2200 *** (Derivatives Use = 1) [A = 732] (4.19) [2] Nonusers of derivatives -0.1395 0.4691 *** (Derivatives Use = 0) [A = 226] (5.14) All Firms -0.0375 0.2763 *** [A = 958] (6.04) [1] to [2] 0.1335 *** (2.65) Panel B. Using Number of Types [1] Firms using both FX and IR 0.0155 0.1501 * (Number of Types = 2) [A = 327] (1.75) [2] Firms using either FX or IR -0.0233 0.2805 *** (Number of Types = 1) [A = 405] (4.17) [3] Firms using no derivatives -0.1395 0.4691 *** (Number of Types = 0) [A = 226] (5.14) [1] to [3] 0.1549 *** (2.71) Panel C. Using Number of Contracts [1] Firms using 4 to 6 contracts 0.0903 0.1496 (Number of Contracts = 4 to 6) [A = 226] (0.72) [2] Firms using 1 to 3 contracts -0.0227 0.2476 *** (Number of Contracts = 1 to 3) [A = 624] (4.57) [3] Firms using no derivatives -0.1395 0.4691 *** Number of Contracts = 0) [A = 226] (5.14) [1] to [3] 0.2298 *** (2.83) Panel D. Using Contracted Dollar Amount [1] Firms using a larger dollar -0.0267 0.2571 ** amount of derivatives [A = 269] (2.54) [2] Firms using a smaller dollar 0.0124 0.1481 ** amount of derivatives [A = 270] (2.22) [3] Firms using no derivatives -0.1395 0.4691 *** [A = 226] (5.14) [1] to [3] 0.1128 * (1.87) *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table VII. Family Firms, Derivatives Use, and Valuation This table provides the cross-sectional regressions of excess value on the family-firm indicator, derivatives-use variables, and the other firm characteristics. The variables are defined as follows. Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to sales. Family is an indicator that takes a value of one if founding family is present and zero otherwise. The Derivatives variable is alternatively one of the following: Derivatives Use is a dummy variable that takes a value of one if a firm uses any kind of derivatives against foreign exchange risk or interest rate risk, and otherwise zero; Number of Types is the number of types of derivatives used (i.e., FX and-or IR), or none and takes values from zero to two; Number of Contracts is the number of different derivative contracts used by the firm and based on the Swaps Monitor's database. It can take values from zero to seven; and Contracted Dollar Amount is the total dollar amount of derivatives used. Number of Types, Number of Contracts, and Contracted Dollar Amount are transformed by adding one and taking the natural log. Size is the log of total assets. Leverage is the ratio of long-term debt to total assets. Note that, t-statistics are shown in parentheses. Dependent Derivatives Variable: Derivatives Use Variable: Excess Value [1] [2] [3] Derivatives 0.1505 *** 0.1501 *** 0.2428 *** (3.13) (3.18) (4.23) Family 0.2518 *** 0.4570 *** (5.73) (5.38) Family * -0.2752 *** Derivatives (-2.82) Size -0.9243 ** -0.8722 ** -0.8948 ** (-2.17) (-2.08) (-2.14) EBIT/Sales 1.2737 *** 1.3344 *** 1.3474 *** (3.38) (3.60) (3.65) Capital 0.1176 0.1606 0.1607 Exp./Sales (0.40) (0.55) (0.55) Size -0.3642 * -0.3875 * -0.3731 * (1 lag) (-1.71) (-1.86) (-1.79) EBIT/Sales 0.8113 * 0.8118 * 0.8327 * (1 lag) (1.86) (1.89) (1.95) Capital Exp./Sales -0.6280 -0.6745 -0.6017 (1 lag) (-0.92) (-1.00) (-0.90) Size 0.0306 0.0538 0.0386 (2 lag) (0.21) (0.38) (0.27) EBIT/Sales 1.3566 *** 1.2579 *** 1.2149 *** (2 lag) (3.61) (3.40) (3.30) Capital Exp./Sales 0.5640 0.4958 0.5445 (2 lag) (1.02) (0.91) (1.01) Leverage -0.2408 -0.1438 -0.1343 (-1.48) (-0.90) -0.84) Size (2) 0.0274 *** 0.0267 *** 0.0272 *** (2.98) (2.96) (3.02) Intercept 13.8769 *** 12.9595 *** 13.1706 *** (3.00) (2.85) (2.90) N 933 933 933 Adj. [R.sup.2] 0.1879 0.2151 0.2210 Dependent Derivatives Variable: Number of Types Variable: Excess Value [4] [5] [6] Derivatives 0.1711 *** 0.1822 *** 0.2746 *** (3.38) (3.66) (4.71) Family 0.2581 *** 0.4577 *** (5.88) (5.76) Family * -0.3161 *** Derivatives (-3.01) Size -0.9030 ** -0.8539 ** -0.8362 ** (-2.12) (-2.04) -2.01) EBIT/Sales 1.2860 *** 1.3495 *** 1.3579 *** (3.42) (3.65) (3.69) Capital 0.0909 0.1318 0.1224 Exp./Sales (0.31) (0.45) (0.42) Size -0.3688 * -0.3959 * -0.3813 * (1 lag) (-1.74) (-1.90) (-1.84) EBIT/Sales 0.7886 * 0.7914 * 0.8093 * (1 lag) (1.81) (1.85) (1.90) Capital Exp./Sales -0.5866 -0.6316 -0.5220 (1 lag) (-0.86) (-0.94) (-0.78) Size 0.0243 0.0476 0.0310 (2 lag) (0.17) (0.34) (0.22) EBIT/Sales 1.3500 *** 1.2413 *** 1.2094 *** (2 lag) (3.60) (3.36) (3.29) Capital Exp./Sales 0.5739 0.5049 0.5407 (2 lag) (1.04) (0.93) (1.00) Leverage -0.2451 -0.1459 -0.1233 (-1.51) (-0.91) (-0.77) Size (2) 0.0270 *** 0.0265 *** 0.0261 *** (2.95) (2.95) (2.91) Intercept 13.8267 *** 12.9818 *** 12.7528 *** (2.99) (2.86) (2.82) N 933 933 933 Adj. [R.sup.2] 0.1893 0.2179 0.2247 Dependent Derivatives Variable: Number of Contracts Variable: Excess Value [7] [8] [9] Derivatives 0.1010 *** 0.1216 *** 0.1746 *** (2.63) (3.21) (4.03) Family 0.2649 *** 0.4185 *** (6.01) (5.51) Family * -0.2037 ** Derivatives (-2.48) Size -0.8231 * -0.7653 * -0.7340 * (-1.93) (-1.83) (-1.76) EBIT/Sales 1.2867 *** 1.3549 *** 1.3629 *** (3.41) (3.66) (3.69) Capital 0.1138 0.1537 0.1488 Exp./Sales (0.38) (0.53) (0.51) Size -0.3558 * -0.3877 * -0.3793 * (1 lag) (-1.67) (-1.86) (-1.82) EBIT/Sales 0.7778 * 0.7831 * 0.7885 * (1 lag) (1.78) (1.83) (1.85) Capital Exp./Sales -0.6349 -0.6854 -0.6023 (1 lag) (-0.93) (-1.02) (-0.90) Size 0.0223 0.0445 0.0357 (2 lag) (0.15) (0.32) (0.25) EBIT/Sales 1.3737 *** 1.2516 *** 1.2431 *** (2 lag) (3.65) (3.39) (3.37) Capital Exp./Sales 0.5801 0.5123 0.5711 (2 lag) (1.05) (0.94) (1.05) Leverage -0.2174 -0.1103 -0.0864 (-1.34) (-0.69) (-0.54) Size (2) 0.0250 *** 0.0244 *** 0.0236 *** (2.73) (2.71) (2.63) Intercept 12.8467 *** 11.9768 *** 11.6022 ** (2.79) (2.64) (2.57) N 933 933 933 Adj. [R.sup.2] 0.1854 0.2153 0.2197 Dependent Derivatives Variable: Contracted Variable: Excess Dollar Amount Value [10] [11] [12] Derivatives 0.0107 *** 0.0111 *** 0.0173 *** (2.88) (3.06) (4.02) Family 0.2583 *** 0.4309 *** (5.64) (5.42) Family * -0.0202 *** Derivatives (-2.65) Size 0.6063 0.6391 0.5685 (1.18) (1.27) (1.13) EBIT/Sales 1.4720 *** 1.5774 *** 1.5846 *** (3.49) (3.82) (3.85) Capital 0.0191 0.0385 0.0434 Exp./Sales (0.07) (0.13) (0.15) Size -0.3061 -0.3379 -0.3217 (1 lag) (-1.43) (-1.62) (-1.54) EBIT/Sales 0.3354 0.3526 0.3620 (1 lag) (0.71) (0.76) (0.78) Capital Exp./Sales -0.5693 -0.6608 -0.5657 (1 lag) (-0.78) (-0.93) (-0.80) Size 0.0361 0.0562 0.0384 (2 lag) (0.25) (0.39) (0.27) EBIT/Sales 1.2292 *** 1.1203 *** 1.0817 *** (2 lag) (3.08) (2.86) (2.78) Capital Exp./Sales 0.6261 0.6489 0.6919 (2 lag) (1.07) (1.14) (1.22) Leverage -0.3179 * -0.1904 -0.1791 (-1.87) (-1.13) (-1.07) Size (2) -0.0088 -0.0088 -0.0072 (-0.77) (-0.79) (-0.65) Intercept -3.5792 -4.1538 -3.3914 (-0.64) (-0.76) (-0.62) N 745 745 745 Adj. [R.sup.2] 0.1789 0.2120 0.2184 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table VIII. Family Firms, Hedging Policies, and Valuation This table presents the cross-sectional regressions of excess value on derivatives use variables for subgroups classified based on family presence (Family) and industrial diversification (Diversification). The variables are defined as follows. Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to sales. Family is an indicator that takes a value of one if founding family is present and zero otherwise. Diversification is an indicator that takes a value of one if the firm reports two or more business segments and zero if it reports only one segment. Derivatives Use is a dummy variable that takes a value one if a firm uses any kind of derivatives against foreign exchange risk or interest rate risk, and otherwise zero. Size is the log of total assets. Leverage is the ratio of long-term debt to total assets. Note that, t-statistics are shown in parentheses. We test coefficient differences between Diversified family * Derivatives Use and Diversified Non-family * Derivatives Use, and report the F-statistic with a p-value in bracket. Dependent Variable: Excess Value Derivatives Use -0.1462 (-1.31) Diversified Family -0.4441 *** (-2.85) Diversified Non-family -0.6615 *** (-5.54) Undiversified Non-family -0.6075 *** (-4.80) Diversified Family * 0.3519 * Derivatives Use (1.95) Diversified Non-family * 0.3724 *** Derivatives Use (2.67) Undiversified Non-family * 0.1703 Derivatives Use (1.11) Size -1.0441 ** (-2.28) EBIT/Sales 2.1027 *** (4.17) Capital Expenditure/Sales 0.8846 (1.07) Size (1 lag) -0.4283 * (-1.82) EBIT/Sales (1 lag) 0.7060 (1.26) Capital Expenditure/Sales (1 lag) -1.5457 (-1.45) Size (2 lag) 0.0711 (0.45) EBIT/Sales (2 lag) 1.5435 *** (3.24) Capital Expenditure/Sales (2 lag) -0.0434 (-0.06) Leverage 0.2763 (1.45) Size (2) 0.0320 *** (3.28) Intercept 15.2079 *** (3.09) Coefficient difference test: Coeff. of Diversified Family * Derivatives Use--Coeff. of 0.02 Diversified Non-family * Derivatives Use [0.9002] N 671 Adj. [R.sup.2] 0.2778 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level. Table IX. Robustness Tests: Alternative Estimation Methods Panel A reports the 2SLS regressions of excess value on operational hedging based on alternative estimation methods. Based on Campa and Kedia (2002), the first step involves a probit model wherein the diversification dummy is estimated. We report the second-stage regression result in this table. Column [1] provides the results using a fixed-effects model. Column [2] reports the results using a random-effects model. Column [3] presents the results using a self-selection model. Column [4] reports the results using standard errors robust to clustering at the firm level and White's (1980) heteroskedasticity. Column [5] provides the results using only the first-year data of each firm. Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995) using the valuation multiple of total capital to sales. Family is an indicator that takes a value of one if founding family is present and zero otherwise. Diversification is an indicator that takes a value of one if the firm reports two or more business segments and zero if it reports only one segment. The controlling variables are the ones used in Table V. Panel B provides the regressions of excess value on financial hedging based on alternative estimation methods. Note that, t-statistics are shown in parentheses. [1] Fixed- [2] Dependent Variable: Effects Random-Effects Excess Value Model Model Panel A. Family Firms, Operational Hedging, and Valuation Family 0.2696 *** 0.3762 *** (2.74) (4.44) Diversification 0.4482 *** 0.4638 *** (4.48) (5.20) Family * -0.3507 *** -0.4501 *** Diversification (-2.59) (-3.70) Controls Yes Yes N 1,450 1,450 R (2) ([dagger]) 0.1728 0.2270 Panel B. Family Firms, Financial Hedging, and Valuation Family 0.0739 ** 0.0943 *** (2.35) (3.07) Derivatives 0.1091 0.2014 *** (1.36) (2.98) Family * -0.1062 ** -0.1154 ** Diversification (-2.03) (-2.23) Controls Yes Yes N 933 933 R (2) ([dagger]) 0.1006 0.2006 [4] Clustering [3] and Hetero. Dependent Variable: Self-Selection Robust Excess Value Model Model Panel A. Family Firms, Operational Hedging, and Valuation Family 0.2842 *** 0.5216 *** (5.74) (3.35) Diversification 0.3398 *** 0.5114 *** (4.35) (2.93) Family * -0.1905 *** -0.5880 *** Diversification (-2.91) (-2.72) Controls Yes Yes N 1,450 1,437 R (2) ([dagger]) 467.5 0.2518 Panel B. Family Firms, Financial Hedging, and Valuation Family 0.3509 ** 0.2428 *** (2.26) (2.89) Derivatives 0.4948 *** 0.4570 *** (5.32) (3.55) Family * -0.2189 ** -0.2752 ** Diversification (-2.03) (-2.09) Controls Yes Yes N 671 933 R (2) ([dagger]) 269.4 0.2327 [5] Dependent Variable: First-Year- Excess Value Only Model Panel A. Family Firms, Operational Hedging, and Valuation Family 0.4702 ** (2.22) Diversification 0.5589 ** (2.16) Family * -0.6105 * Diversification (-1.84) Controls Yes N 208 R (2) ([dagger]) 0.3240 Panel B. Family Firms, Financial Hedging, and Valuation Family 0.2365 * (1.79) Derivatives 0.3213 * (1.66) Family * -0.0851 Diversification (-0.39) Controls Yes N 205 R (2) ([dagger]) 0.2727 *** Significant at the 0.01 level. ** Significant at the 0.05 level. Significant at the 0.10 level. ([dagger]) Wald [chi square] is reported for the self-selection model. Table X. Robustness Tests: Using 2004 Data This table retests the hedging effects on excess value using 2004 data. Family firms are obtained from the November 10, 2003 issue of Business Week. Panel A compares a firm's Excess Value of subgroups replicating Table VI. it reports mean differences between firms with comprehensive- sophisticated contracts and nonusers of derivatives. Excess Value is the measure of a firm's excess value based on Berger and Ofek (1995). Derivatives Use is a dummy variable that takes a value one if a firm uses any kind of derivatives against foreign exchange risk or interest rate risk, and otherwise zero. Number of Types is the number of types of derivatives used (i.e., FX and/or IR), or none and takes values from zero to two. Number of Contracts is the number of different derivative contracts used by the firm and is based on the Swaps Monitor's database taking values from zero to seven. Contracted Dollar Amount is the total dollar amount of derivatives used. Panel B replicates Table VII reporting the cross-sectional regressions of excess value on the family-firm indicator, derivatives--use variables, and the other firm characteristics. Panel A. Comparisons Non-family of Excess Value Family Firms Firms All Firms (Family = 1) (Family = 0) Panel A.1: Using Derivatives Use Users of derivatives -0.3598 *** 0.0510 -0.0958 minus nonusers of (-3.57) (0.67) (-1.55) derivatives Panel A.2: Using Number of Types -0.3542 *** -0.0200 -0.1384 * Firms using both FX (-3.16) (-0.22) (-1.95) and IR minus firms using no derivatives Panel A.3: Using Number of Contracts 0.2644 0.0833 -0.0341 Firms using 2 to (1.20) (0.51) (-0.26) 4 contracts minus firms using no derivatives Panel A.4: Using Contracted Dollar Amount Firms using a larger -0.4155 ** 0.0970 -0.0454 dollar amount of (-2.30) (0.87) (-0.47) derivatives minus firms using no derivatives Panel B. Regression of Excess Value on Family Firm and Derivatives Use Derivatives Derivatives Variable: Variable: Derivatives Number Use of Types [1] [2] Derivatives 0.0321 -0.0237 (0.45) (-0.32) Family 0.1255 0.0730 (1.30) (0.79) Family * Derivatives -0.2782 ** -0.2128 * (-2.31) (-1.73) Controls Yes Yes N 338 338 Adj. [R.sup.2] 0.1798 0.1776 Derivatives Derivatives Variable: Variable: Number Contracted of Contracts Dollar Amount [3] [4] Derivatives 0.0313 0.0030 (0.34) (0.65) Family -0.0188 0.0411 (-0.27) (0.60) Family * Derivatives -0.1273 -0.0231 *** (-0.83) (-2.81) Controls Yes Yes N 338 338 Adj. [R.sup.2] 0.1646 0.1857 *** Significant at the 0.01 level. ** Significant at the 0.05 level. * Significant at the 0.10 level.

Printer friendly Cite/link Email Feedback | |

Author: | Kim, Chansog; Pantzalis, Christos; Park, Jung Chul |
---|---|

Publication: | Financial Management |

Date: | Jun 22, 2014 |

Words: | 17398 |

Previous Article: | Bucking the trend: the informativeness of analyst contrarian recommendations. |

Next Article: | When does a merger create value? Using option prices to elicit market beliefs. |

Topics: |