The Association between Financial Reporting Quality and Investment Decisions for Family Firms in an Emerging Market

Transcription

1 The Association between Financial Reporting Quality and Investment Decisions for Family Firms in an Emerging Market Chan-Jane Lin Department and Graduate Institute of Accounting National Taiwan University Taipei, 106 Taiwan Tawei Wang School of Accountancy Shidler College of Business University of Hawaii at Manoa Honolulu, HI Chao-Jung Pan Department and Graduate Institute of Accounting National Taiwan University Taipei, 106 Taiwan Corresponding author

2 The Association between Financial Reporting Quality and Investment Decisions for Family Firms in an Emerging Market Abstract This paper investigates the relation between financial reporting quality and investment inefficiency for family firms in an emerging market. Building on prior literature regarding family firms financial reporting quality and investment behavior, we hypothesize that family firms are more likely to under-invest and financial reporting quality can help alleviate such behavior. Using a sample of listed firms in Taiwan from 1996 to 2009, we show that, comparing to non-family firms, financial reporting quality can reduce family firms inefficient investment behavior, especially for under-investment. In addition, for family firms, when the deviation of control from ownership and uncertainty are high, such effect is more pronounced. Keywords: investment inefficiency; financial reporting quality; family firm JEL Classifications: M41, G31, D81 1

3 The Association between Financial Reporting Quality and Investment Decisions for Family Firms in an Emerging Market 1. Introduction Prior literature has documented large shareholders impact on a firm s investment decisions. For example, Fama and Jensen (1985) state that large, undiversified shareholders may invest based on their own risk preferences. Differently, Edmans (2009) shows that blockholders can reduce managerial incentives to pursue myopic investment decisions. Our study also focuses on investment decisions but on a particular class of large shareholders, family owners. Family controlled firms are prevalent in the world, especially in Europe and in Asia (Faccio and Lang, 2002; Anderson and Reeb, 2003), and have contributed a significant portion to the world s GDP (Fan et al., 2011). Family firms investment decisions, different from non-family firms, are shaped by the commitment of the family (Carlock, 2010). In addition, as family owners interests are linked tightly to the firm s performance, family owners need to make sure not to endanger the existing family business and the long-term survival of the firm when making investment decisions (e.g., Carlock, 2010; Klein and Ward, 2012). The above mentioned factors raise the concern of investment inefficiency of family firms. Investment inefficiency is defined as a firm gives up investment projects that can otherwise increase its value or invest in projects that do not increase its value (e.g., Biddle et al., 2009). A number of recent studies have investigated the relation between financial reporting quality and investing decisions (e.g. Bushman et al., 2006; McNichols and Stubben, 2008; Li and Tang, 2008; Biddle et al., 2009; Kedia and Philippon, 2009). These studies claim that, different from other governance mechanisms, financial reporting quality has important economic implications for investment decisions by reducing the agency conflicts between 2

4 managers and external capital suppliers. 1 However, the association between financial reporting quality and investment inefficiency is not ex ante clear in the context of family firms in the emerging market because of the following reasons and as detailed in Section 2. First, based on prior literature, it is inconclusive whether family firms have better financial reporting quality or invest more efficiently, comparing to non-family firms (Anderson and Reeb, 2003; Wang, 2006; Ali et al., 2007; Chen et al., 2008; Anderson et al., 2009; Anderson et al., 2010). Second, prior studies on family firms investment behavior are mainly based on U.S. and European sample firms. The family firms in the U.S. and Europe may behave differently comparing to those in emerging markets (Fan et al., 2011) especially when emerging markets are characterized with lower relevance of accounting information, possibly worse financial reporting quality but fewer information sources other than financial reports (Ali and Hwang, 2000; Ball et al., 2000; Chen et al., 2011). Based on the above discussion, the main purpose of this study is to investigate the impact of financial reporting quality on investment decisions for family versus non-family firms in an emerging market. In particular, as family firms in the emerging market are more likely to under-invest (see detailed discussion in Section 2), we focus on whether financial reporting quality can reduce under-investment behavior more for family firms, comparing to non-family firms. We further examine how the characteristics of family firms, namely, deviation of control from ownership and uncertainty, would affect such association as these two can worsen the agency conflicts between the controlling owner and minority shareholders. To address our research questions, we use the listed firms in Taiwan between 1996 and 2009 as our sample. The Taiwanese firms are selected for our analyses because (1) Taiwan is an emerging market, and (2) approximately two-thirds of the listed firms in Taiwan 1 For example, Biddle et al. (2009) provide mixed evidence on whether alternative governance mechanisms such as institutional ownership and analyst coverage improve investment inefficiency. However, the association between financial reporting quality and investment remains negative and significant after controlling for these corporate governance variables. 3

5 are family firms and the ownership concentration is higher than that of Asian average (Chen and Ho, 2009). Our results suggest that in general, financial reporting quality can enhance investment efficiency. However, comparing to non-family firms, financial reporting quality can improve under-investment more for family firms. Furthermore, under the case of high deviation of control from ownership and high uncertainty, again, financial reporting quality can reduce under-investment more for family firms, comparing to non-family firms. This study contributes to the literature in the following ways. First, our findings contribute to the emerging literature on the relation between financial reporting quality and investment decisions in the context of family firms in an emerging market. In particular, prior studies, such as Biddle et al. (2009), investigate the association between financial reporting quality and investment decisions in the context of diversified ownership and developed market. In a similar context, Verdi (2006) shows that, financial reporting quality is more strongly associated with over-investment for firms with large cash balances and dispersed ownership. However, he finds mixed evidence for the effect on under-investment for firms facing financing constraints. He also shows that the relation is stronger for firms with low analyst following and high bid-ask spread. Our paper highlights the unique characteristics of family firms, comparing to non-family firms. Different from Verdi (2006), our results do not show significant relation between financial reporting quality and investment (both under- and over-investment) for non-family firms (the firms with diversified ownership as in Verdi, 2006). Instead, our findings suggest that financial reporting quality mitigates under-investment only for family firms. Our findings also demonstrate that though financial reporting quality in the emerging market is generally worse comparing to that in developed market, it can still reduce under-investment especially for family firms. Second, our study contributes to the literature in family firms by examining the impact of a monitoring mechanism, financial reporting quality, on investment decisions. Prior studies 4

6 focus more on investigating the consequence of family control such as corporate governance, financial reporting quality and investment or financing strategies (e.g., Ali et al., 2007; Anderson and Reeb, 2003; Chen and Ho, 2009; Wang, 2006; Wei and Zhang, 2008; Yeh et al., 2001; Morck and Yeung, 2003). Our paper investigates not only the implication of family control on the association between financial reporting quality and investment efficiency but also how this implication would change with the deviation of control from ownership and uncertainty. The remainder of the paper is organized as follows. In Section 2, we elaborate on our research background and develop hypotheses. The research methodology is discussed in Section 3. In Section 4, we present the empirical results and robustness tests. We conclude in Section Research Background and Hypothesis Development Financial reporting information can help investors assess the amount, timing, and uncertainty regarding a firm s cash flows. It can also facilitate the efficiency of capital allocation such as a firm s investment decisions (Levine, 1997). From this perspective, higher financial reporting quality can alleviate a firm s investment inefficiency by mitigating information asymmetries between managers and investors (e.g., Leuz and Verrecchia, 2000; Verrecchia, 2001; Bushman and Smith, 2001; Biddle and Hilary, 2006; Biddle et al., 2009; Verdi, 2006; Chen et al., 2011). The reduced information asymmetry could lower the costs of raising funds and/or the costs of monitoring that lead to a more efficient investment level (Verdi, 2006). For example, Biddle and Hilary (2006) document that higher accounting quality reduces investment-cash flow sensitivity (a proxy for investment inefficiency). This effect is stronger in economies where the stock market is the dominant source of capital. Verdi (2006) shows that financial reporting quality is more strongly associated with over-investment for firms with large cash balances or dispersed ownership but finds mixed 5

7 evidence for the relation between financial reporting quality and under-investment for firms facing financing constraints. Biddle et al. (2009) demonstrate that financial reporting quality is negatively associated with both under-investment and over-investment. We examine the role played by financial reporting quality on a firm s investment decisions in the setting of family firms in an emerging market. This setting is important and can provide additional insights to our understanding of the association between financial reporting quality and investment efficiency for the following reasons. First, family-controlled firms are prevalent in eleven of the thirteen continental European countries (Faccio and Lang, 2002), account for one-third of the S&P 500 firms (Anderson and Reeb, 2003) and more than two-thirds of the firms in East Asia (Claessens et al., 2002). In addition, the market capitalization of family firms is about 50% and 27% of nominal GDP in South and North Asia, respectively (Fan et al., 2011). Given the importance of a firm s investment decisions on its operations and family-controlled ownership in the world economy, the critical role played by family firms investment decisions and investment efficiency is evident. Second, financial reporting quality is different between family and non-family firms, but the direction is unclear based on theoretical arguments or empirical evidence. Theoretically, the entrenchment hypothesis states that firms with a centralized ownership are more likely to derive personal interests at the expense of minority shareholders, which may lead to less transparent financial reports. On the other hand, the interest-alignment hypothesis argues that controlling family owners reduce owner-manager conflicts by effectively monitoring professional executives (Anderson and Reeb, 2003), which may also result in less transparent financial reports because shareholders do not need to monitor management. It is also possible that family firms would have better disclosure and transparency due to the better monitoring function. Empirically, the results are also inconclusive with regard to the 6

8 association between family firms and financial reporting quality. Ali et al. (2007) and Wang (2006) find that family firms in the S&P 500 Index have better financial reporting quality. However, Chen et al. (2008) and Anderson et al. (2009) use different samples from S&P 500 firms and show that family firms are less transparent. Evidence using the sample from Asian family firms also find mixed results (e.g., Claessens et al., 2000; Fan and Wong, 2002; Ball et al., 2003). The above observation seems to make the association between financial reporting quality and investment inefficiency as argued in prior literature less obvious in the family firms context. Nevertheless, Verdi (2006) shows that the impact of financial reporting quality on investment efficiency is more important when financial reporting information is the only information source. Unlike their counterparts in Western countries where more information sources, such as financial analyst forecasts, are available, market participants in emerging markets rely more on financial reports to reduce information asymmetry. This makes us believe that financial reporting still plays a critical role in affecting investment decisions even when family firms may have worse financial reporting quality, comparing to non-family firms. Third, it is not ex ante clear either about family firms investment behavior comparing to non-family firms. There are two arguments that may lead to different expectations of the investment strategy between family firms and non-family firms: long-term horizon and risk aversion (Anderson et al., 2010). The long-term horizon argument states that the concentrated ownership of family firms makes it possible that family firms focus more on long-term investment decisions. Specifically, consistent with the interest-alignment hypothesis, the long-term horizon argument also suggests that being continuously committed shareholders, family owners have incentives to make strategic investments that ensure the long-term viability and health of the firm (Anderson and Reeb, 2003; Chen et al., 2008). From this perspective, family firms are less likely to show myopic investment behavior 7

9 (Shleifer and Vishny, 1986; Stein, 1989; James, 1999; Edmans, 2009), and have lower investment-cash flow sensitivities (Galeotti et al., 1994; Wei and Zhang, 2008; Pindado et al., 2011). Based on the above studies, there are reasons to believe that family firms are less likely to have inefficient investment decisions relative to non-family firms. Differently, the risk aversion argument states that family owners have a stronger incentive to lower the firm risk through various activities such as investment strategies (Anderson et al., 2010) because the well-being of the family owners ties more closely to the firm performance (Anderson and Reeb, 2003). Consistent with the entrenchment hypothesis where the agency conflicts between minority and controlling shareholders may result in sub-optimal investments (Fama and Jensen, 1985), the risk aversion argument provides similar expectations. In particular, family firms may allocate fewer financial resources on long-term investments as risks increases with investment levels (Anderson et al., 2010), which suggests that, comparing to non-family firms, family firms are more likely to under-invest. Prior studies on family firms investment behavior, however, are mainly based on U.S. and European sample firms, and the empirical results are mixed. For example, Anderson et al. (2010) find that family firms in the U.S. spend fewer financial resources on long-term investments than non-family firms. Pindado et al. (2011) show that family firms in nine European countries have lower investment-cash flow sensitivities, and invest more efficiently. Similarly, Bjuggren et al. (2008) and Eklund et al. (2011) demonstrate that Swedish family firms have better investment performance than non-family firms. Interestingly, family firms in the U.S. and Europe are in their fourth or even fifth generation whereas most of the family firms in Asia are still in their first generation (Fan et al., 2011). We examine the investment behavior of family firms in Taiwan, hoping to gain some insights into the relevance of financial reporting quality for investment efficiency in an emerging market which is 8

10 characterized with the attributes of weaker investor protection and lower relevance of accounting information (Ball et al., 2000; Ali and Hwang, 2000). As illustrated in the Asian Family Business Report 2011 by Credit Suisse (Fan et al., 2011), listed family firms in Taiwan account for 49% of market capitalization, have consistently delivered superior ROE relative to the market average over the last decade, which is supported by the slight outperformance of stock return against the market since Taiwan is one of the leading suppliers of IT products in the world, and its listed family firms dominate in IT industry, providing two-thirds of job opportunities for that sector. In spite of these, the report indicates that while family firms fixed assets investment stays quite stable, non-family firms contribute to most of the fixed assets investments in Taiwan. In other words, relatively, family firms seem to have under-investment problem over the years. One recent work by Lin et al. (2012) also finds that family firms in Taiwan show more severe under-investment behavior when facing capital shortages. Collectively, the above observations are more in line with the risk aversion argument discussed earlier. Therefore, we argue that family firms in Taiwan would show under-investment behavior. Following the arguments above and those in recent papers about financial reporting quality and investment efficiency, we expect to see that, comparing to non-family firms, financial reporting quality can reduce under-investment more for family-firms. In particular, the financial reporting quality can reduce information asymmetry and serve as a monitoring function that makes the family owners invest more efficiently. Note that, though our focus is on under-investment, in our analysis, we still take over-investment into account. Formally, our first hypothesis is as follow: H1: The effect of financial reporting quality on reducing under-investment for family firms is stronger than that for non-family firms. 9

11 However, the above discussion does not take into consideration the following two main characteristics of family firms that would result in the agency conflict of controlling owner and minority shareholders: the deviation of control from ownership and uncertainty. Prior literature suggests that as the deviation of control from ownership becomes larger, the agency conflict between the controlling owner and minority shareholders is more severe (e.g., Fan and Wong, 2002; Claessens et al., 2002; Baek et al., 2004). Empirical findings indicate that when the conflict is more severe, family firms are more likely to expropriate assets by distributing special dividends (DeAngelo and DeAngelo, 2000), adopting a disproportionate share of earnings (Gilson and Gordon, 2003) and engaging in related-party transactions (Gilson and Gordon, 2003; Anderson and Reeb, 2003; Ali et al., 2007). Specific to investment decisions, when the level of separation of ownership and control is higher, family firms may reject innovation to protect their obsolete investment (Morck and Yeung, 2003), which leads to under-investment. A couple recent papers show similar results. Specifically, Bjuggren and Palmberg (2010) find that dual-class share (a proxy for the deviation of control from ownership) has a significant negative impact on investment performance. Pindado et al. (2011) suggest that family firms with control-enhancing mechanisms have higher investment-cash flow sensitivity than family firms without such mechanisms. The above two studies suggest that the separation of control and ownership would lead to investment inefficiency. Eklund et al. (2011) demonstrate that the negative impact of the deviation on investment performance is larger for family firms than for non-family firms. In line with these findings, compared to non-family firms, this negative impact of deviation on under-investment is more serious for family firms. We expect that, as the deviation of control and ownership increases, (1) for family firms, financial reporting quality can deter under-investment more, and (2) comparing to non-family firms, financial reporting 10

12 quality has a larger impact on under-investment for family firms. Specifically, our second set of hypotheses is as follows: H2a: Financial reporting quality reduces under-investment more for family firms with high deviation of control from ownership as opposed to those with low deviation. H2b: For firms with high deviation of control from ownership, financial reporting quality reduces under-investment more for family firms than that for non-family firms. The second characteristic that Hypothesis 1 does not consider is uncertainty. When firms have greater information asymmetry, as family firms do, it is difficult and costly for outsiders to evaluate managers investment and overall performance (Francis and Martin, 2010), which may result in investment inefficiency. In addition, the investment behavior of family firms seems to be more sensitive to uncertainty because of the irreversibility of their investment decisions and their risk-aversion tendency (Bianco et al., 2009). Family firms are also slower when responding to new opportunities (Cucculelli, 2008). Anderson et al. (2010) find that family ownership is negatively related to total investment (i.e., under-investment) when uncertainty is high. From the above discussion, we believe that the negative impact of uncertainty on under-investment is possibly more serious for family firms than it is for non-family firms especially in the emerging market. This is because, again, there are fewer sources other than financial reports in the emerging market for the investors to obtain information about a firm (Chen et al., 2011). Specifically, we expect that, as the uncertainty increases, (1) for family firms, financial reporting quality can deter under-investment more, and (2) comparing to non-family firms, financial reporting quality has a larger impact on under-investment for family firms. Our third set of hypotheses is as follows: 11

13 H3a: Financial reporting quality reduces under-investment more for family firms with high uncertainty as opposed to those with low uncertainty. H3b: For firms with high uncertainty, financial reporting quality reduces under-investment more for family firms than that for non-family firms. 3. Research Methodology 3.1. Empirical Model and Measures We use Equation (1) to test our hypotheses. INV i,t+1 = β 0 + β 1 FQ i,t + β 2 RANKFCF_FQ + β 3 RANKFCF i,t+1 + β j CONTROL j,i,t + ε i,t+1 (1) INV i,t+1 is defined as the sum of research and development expenditure, capital expenditure, and acquisition expenditure, minus cash receipts from sale of fixed assets, and deflated by lagged total assets for firm i at time t+1 (see Appendix for variable definitions). FQ i,t is the measure of financial reporting quality for firm i at time t. Consistent with prior literature (e.g., Aboody et al., 2005; Francis et al., 2005; Verdi, 2006; Biddle et al., 2009), FQ is the standard deviation of the firm-level residuals based on the modified Dechow and Dichev (2002) model. To estimate FQ, we require all the industries to have at least 9 observations in a given year. Following Biddle et al. (2009), RANKFCF i,t+1 is the likelihood or tendency of inefficient investment for firm i at time t+1, which is calculated by ranking the free cash flows in deciles by industry-year and re-scaling between 0 and 1. As firms with free cash flows are more likely to engage in sub-optimal investment (Jensen, 1986; Stulz, 1990; Richardson, 2006), we use it to capture when over- or under-investment is more likely to happen. RANKFCF_FQ is the interaction term of FQ i,t and RANKFCF i,t+1. The coefficient of FQ captures the effect of FQ on investment inefficiency when firms are more likely to under-invest, while the coefficient of RANKFCF_FQ captures the marginal effect when firms are more likely to over-invest. The summation of the coefficients of FQ and 12

14 RANKFCF_FQ represents the effect of FQ on investment inefficiency when firms are most likely to over-invest (i.e., when RANKFCF equals one). CONTROL i,t is a set of control variables for firm i at time t widely used in prior literature to control for the determinants of investment activities (e.g., Biddle and Hilary, 2006; Richardson, 2006; Biddle et al., 2009). We include four corporate governance variables: institutional ownership (INSTITUTION, the percentage of a firm s shares held by institutional investors), stocks pledged by directors and supervisors (PLEDGE, the percentage of the stocks held by directors and supervisors being pledged), the proportion of independent directors and supervisors (INDBOARD, the percentage of independent directors and supervisors on the board) and CEO duality (DUALITY, equals 1 if the CEO is also the chairman; 0 otherwise). In addition, we control for firm characteristics that could affect investment decisions for firm i at time t. These characteristics are (1) growth of the firm (Q, the book value of total debt plus the market value of equity divided by the book value of total assets), (2) firm size (SIZE, the natural log of total assets), (3) leverage of the firm (LEVERAGE, the sum of the book value of short-term and long-term debts deflated by the sum of the book value of total debts and equity), (4) liquidity (LIQUIDITY, the balance of cash and short term investments deflated by lagged total assets.), (5) age of the firm (AGE, the natural log of the difference between the year when the firm was established and the current year), (6) stock return (STOCKRETURN, the change in market value of the firm in the current year), (7) the investment amount in prior period (LAGINV), and whether a firm s financial reports are audited by big 4 CPA firms (BIG4, equals 1 if a firm s financial reports are audited by a big 4 CPA firm; 0 otherwise). We also take into account the industry and year fixed effects in our model. The regression model is estimated using OLS with Huber-White standard errors. According to our hypotheses, it is expected that the coefficient on reporting quality is greater than zero (i.e., β 1 > 0) given financial reporting quality can reduce the level of 13

15 under-investment. We adopt two definitions for family firms based on Lin and Chang (2009) for H1. First, FAMILY1 equals 1 if half (or higher) of the board seats is controlled by the ultimate owner via individual or the affiliated firms (e.g., listed companies and non-listed companies he/she controls), 0 otherwise. Second, FAMILY2 equals 1 if the ultimate owner meets the requirement of FAMILY1 definition and owns more than 10 percent of the firm s shares. For H1, we expect that β 1 of the family firm sub-sample is significantly positive and larger than that of the non-family firm sub-sample. For the second and the third hypothesis, we use the following measures for deviation of control from ownership and uncertainty. First, for deviation of control from ownership (DEVIATION), we use the standard measure in prior studies (e.g., Fan and Wong, 2002; Claessens et al., 2002), which is the difference between voting rights and cash flow rights possessed by the largest ultimate shareholder of the firm. Details of the calculation of this measure are given in Appendix. Second, for uncertainty, we construct a composite index, UNCERTAINTY, to capture it. Specifically, the uncertainty faced by firm i at time t is an index formulated by three variables: (1) VOLATILITY i,t : the standard deviation of daily stock returns for firm i at time t, (2) BAS i,t : the bid-ask-spread defined for firm i at time t as the annual average of daily spread scaled by the midpoint between bid and ask, and (3) VOLUME i,t : the annual average of daily trading volume for firm i at time t. To test our second hypothesis, we rank DEVIATION and UNCERTAINTY into terciles. For H2a (H3a), we expect that, for family firms, β 1 is positive and larger in the high deviation (uncertainty) group than that in the low deviation (uncertainty) group. For H2b (H3b), we expect that, for high deviation (uncertainty) firms, β 1 is positive and larger for the family firms than that for non-family firms. 14

16 3.2. Sample and Descriptive Statistics We collect all the listed firms except financial institutions in Taiwan from 1996 to 2009 from the Taiwan Economic Journal (TEJ) database. The variables mentioned in Section 3.1 are also gathered from or calculated based on the information provided in the TEJ database. Our initial sample consists of 16,987 firm-year observations. We exclude the observations with missing values and a firm with non-calendar fiscal year end. We also delete the top and bottom 1% of the observations for INV and FQ. The resulting sample consists of 6,528 firm-year observations. The year and industry breakdown of our sample are given in Panel A and Panel B of Table 1, respectively. As shown in Table 1 Panel A, we have more observations in recent years due to the economic growth and the development of capital market. Panel B in Table 1 demonstrates that our sample distributes in 14 different industries with about 46% in the electronics industry. Panel C in Table 1 presents the descriptive statistics for the entire sample. On average, firms invest 4.3 percent the value of prior years assets (the mean of INV). Also, about 33 percent, on average, of outstanding shares are owned by institutions (the mean of INSTITUTION) and about 12 percent of the stocks held by directors and supervisors are pledged (the mean of PLEDGE). Panel D in Table 1 shows the descriptive statistics for the sub-sample of family and non-family firms based on the definition of FAMILY2. As shown in Panel D, about 67 percent (4,349/6,528) of the firms in our sample are family firms while about 33 percent are non-family firms (2,179/6,528). When comparing the means of all the variables between family and non-family firms, they are all significantly different (p < 0.01) except for UNCERTAINTY. In particular, on average, comparing to non-family firms, family firms invest significantly less (the mean of INV: < 0.051, p < 0.01) which is consistent with the Credit Suisse research report mentioned earlier (Fan et al., 2011) and prior literature (e.g., Lin et al., 2012). Furthermore, family firms have better financial reporting 15

17 quality (the mean of FQ for family firms, , is larger than that for non-family firms, , p < 0.01), have higher deviation of control from ownership (DEVIATION), are larger in terms of total assets (SIZE) and have longer years in business (AGE). (Insert Table 1 about here) The Pearson correlation is given in Table 2. It shows that financial reporting quality (FQ) is significantly and negatively associated with investment (INV). Furthermore, investment increases with growth opportunity (Q), cash and short-term investment percentage in total assets (LIQUIDITY) and stock returns (STOCKRETURN) but decreases with the age of the firm (AGE) and leverage (LEVERAGE). However, we do not observe high correlations that raise multicollinearity concerns (Insert Table 2 about here) 4. Empirical Results 4.1. Main Results Table 3 presents the results of our empirical model based on the full sample, family and non-family firm sub-samples. First, for the full sample, our results show that the coefficient of FQ is significantly positive (0.111, p < 0.01), indicating that a higher financial reporting quality can enhance investment efficiency when firms are more likely to under-invest. The results also show an incremental effect that higher financial reporting quality can mitigate the over-investment behavior when firms tend to over-invest, i.e., the coefficient of RANKFCF_FQ is significantly negative. However, when firms are most likely to over-invest (i.e., RANKFCF is one), such association is insignificant (i.e., the summation of 16

18 the coefficients of FQ and RANKFCF_FQ is insignificant). Second, the results for the family and non-family firm sub-sample are reported based on the two definitions of family firms, namely, FAMILY1 and FAMILY2. The results under these two definitions of family firms are similar. Specifically, for family firms, when firms are more likely to under-invest, financial reporting quality is more positively associated with investment (the coefficients of FQ are and 0.153, p < 0.01, respectively). In addition, the impact of financial reporting on over-investment, again, is only marginal (the coefficients of RANKFCF_FQ are significantly negative) but insignificant when firms are most likely to over-invest (the summation of the coefficients of RANKFCF_FQ and FQ is insignificant). We then compare the results for family firms with those for non-family firms. Consistent with our hypothesis, we only observe the significant mitigating effect of financial reporting quality on under-investment for family firms. For non-family firms, the associations between financial reporting quality and investment are all insignificant (both under- and over-investment). This result suggests that as family firms, comparing to non-family firms, are more conservative in making long-term investment decisions, financial reporting quality can help improve investment efficiency. (Insert Table 3 about here) We further investigate whether the above results are affected by different levels of deviation of control from ownership (Table 4) as well as uncertainty (Table 5). Table 4 presents the results for family/non-family firms and high/low (i.e., top/bottom tercile) deviation of control from ownership. We first compare the results for family firms given different deviation levels. We find that for Family Firms_High Deviation group, financial reporting quality can improve investment efficiency (the coefficient of FQ is 0.198, p < 0.05) 17

19 when a firm is more likely to under-invest. However, we do not observe similar association for Family Firms_Low Deviation group, which supports H2a that financial reporting quality reduces under-investment more for family firms with high deviation of control from ownership. Then we compare the results for family and non-family firms given the high/low deviation group. For the high deviation group, the above mentioned finding is only true for family firms but not for non-family firms. This is consistent with H2b that family firms are manifestations of high deviation of control from ownership and such characteristics make the impact of financial reporting quality on investment inefficiency more pronounced. For the low deviation group, as expected from prior literature (e.g., Biddle et al., 2009) of a firm in general (i.e., low deviation and dispersed ownership), we observe the association between financial reporting quality and investment only for the non-family firm group (0.160, p < 0.10) when a firm is more likely to under-invest. We have similar findings for the family/non-family and high/low uncertainty group as given in Table 5. In particular, for the Family Firms_High Uncertainty group, when a firm is more likely to under-invest, financial reporting quality is significantly and positively associated with investment (0.274, p < 0.05). In addition, financial reporting quality has an incremental effect on investment efficiency when a firm is more likely to over-invest (the coefficient of RANKFCF_FQ is , p < 0.10). However, this is not the case for the Family Firms_Low Uncertainty group or the Non-Family Firms_High Uncertainty group. This is consistent with H3a and H3b that as family firms, comparing to non-family firms, are more risk-averse and are slow in responding to the market, the impact of financial reporting quality on investment inefficiency is more profound with high uncertainty. For the Non-Family Firms_Low Uncertainty group, again, consistent with prior literature (e.g., Biddle et al., 2009) for a firm in general (i.e., low uncertainty and dispersed ownership), financial reporting quality can improve investment efficiency when a firm is more likely to 18

20 under-invest (0.618, p < 0.05). (Insert Table 4 and Table 5 about here) In conclusion, our results show that financial reporting quality can reduce investment inefficiency for family firms more than for non-family firms, especially for under-investment. In addition, with higher level of deviation of control from ownership and uncertainty, such impact is stronger for family firms, comparing to non-family firms Robustness Tests We perform the following analyses to further validate our empirical results. First, the acquisition expenditure in the TEJ database is the sum of cash payments and cash receipts for acquisition activities. This net value may cause measurement errors in investments. To avoid such error, investment is recomputed as the sum of research and development expenditure and capital expenditure minus cash receipts from sale of property, plant and equipment divided by lagged total assets. Our results are qualitatively similar based on this measure. Next, we consider the interaction effect between corporate governance variables and the likelihood of inefficient investment in the investment model as in prior studies (e.g., Biddle et al., 2009). Our results are largely similar. Third, we use the model in Richardson (2006) to estimate the amount of under- and over-investment. That is, we first regress the investment amount (INV) on Q, SIZE, AGE, LEVERAGE, LIQUIDITY, STOCKRETURN, and LAGINV. The positive (negative) residuals from the model are called over-investment (under-investment). We then separately regress the absolute value of under- and over-investment on financial reporting quality (FQ). Our results are largely similar. Fourth, since we are using panel data, we further control for the potential serial 19

21 correlation as suggested by Petersen (2009). Our results remain similar when using clustered standard errors by firm-year. Last, we take into account the issue that financial reporting quality could potentially be an endogenous variable. To address this issue, we use a two-stage least squares (2SLS) model. Following the steps suggested by Larcker and Rusticus (2010), we first identify two instrument variables that would affect financial reporting quality but least likely to affect investment decisions and correlate with the residuals in Equation (1). These two variables, used as control variables in related studies (Prawitt et al., 2009; Cao et al., 2012), are return on assets (ROA) and a dummy variable indicating whether a firm has operating loss (LOSS). Then we include the instruments and all the independent variables in the first stage model to check for weak instruments. The results are given in Table 6. Note that for readers convenience, we attach our main results in Table 3 in the second column for references. From the results of the first-stage model, our acceptable partial R 2 and high F-statistics suggest that our instruments are not weak. In addition, since we use multiple (two) instruments, we also make sure that we do not have over-identifying issues (χ 2 = and 1.649, n.s.). The Hausman test also supports that the two-stage model is more preferable for the family firms sub-sample. Table 6 shows that our second stage results (the right-most column) are similar to our main results in Table 3, suggesting that our findings are robust after taking into account the endogeneity of financial reporting quality. Unreported results for the two-stage model for deviation of control from ownership and uncertainty are also similar to those in Table 4 and 5. (Insert Table 6 about here) 20

22 5. Conclusions Family firms play an important role in today s world economy as mentioned in Introduction. The efficiency of their investment decision would profoundly affect the economic development especially in the emerging market. Given the importance of their economy impact and, comparing to non-family firms, their inconclusive financial reporting quality and investment behavior from prior literature, our paper focuses on the function of financial reporting quality on deterring investment inefficiency, especially under-investment, given family versus non-family firms. Our results suggest that, consistent with prior literature, higher financial reporting quality indeed improves investment efficiency. However, comparing to non-family firms, such association only exists for family firms and in the case of under-investment. This result also holds when taking the deviation of control from ownership and uncertainty into account. In particular, the mitigating effect of financial reporting quality on under-investment is more pronounced for family firms and under high deviation of control from ownership and uncertainty. Our paper contributes to the emerging literature regarding the association between financial reporting quality and investment decisions. Different from prior literature, such as Verdi (2006) and Biddle et al. (2009), our findings mentioned above demonstrate the uniqueness of family firms in the emerging market. More importantly, though financial reporting quality in the emerging market is often considered worse than that in developed market, it can still play its role in affecting investment decisions. While our results indicate that higher financial reporting quality can mitigate family firms under-investment behavior, some caveats are as follows. First, though we hypothesize and find that family firms are more conservative in making investment decisions, family firms risk preference and risk avoiding behavior are not captured in our analyses. Second, we do not consider the generation of the family owner, such as founder or heirs. 21

23 The generation can also affect a firm s investment decisions. However, as most of the family owners in Taiwan are still in their first generation, this factor is less likely to affect our results. Last, there are other possible ways to proxy or measure financial reporting quality and investment efficiency. Each of these measures (including ours) has potential measurement errors. 22

24 References Aboody, D., Hughes, J. and Liu, J. (2005) Earnings quality, insider trading, and cost of capital, Journal of Accounting Research, 43, pp Ali, A. and Hwang, L. S. (2000) Country-specific factors related to financial reporting and the value relevance of accounting data, Journal of Accounting Research, 38 (1), pp Ali, A., Chen, T. and Radhakrishnan, S. (2007) Corporate disclosures by family firms, Journal of Accounting and Economics, 44, pp Amihud, Y. and Lev, B. (1981) Risk reduction as a managerial motive for conglomerate mergers, The Bell Journal of Economics, 12, pp Anderson, R. and Reeb, D. (2003) Founding-family ownership and firm performance: Evidence from the S&P 500, Journal of Finance, 58, pp Anderson, R., Duru, A. and Reeb, D. (2009) Founds, heirs and corporate opacity, Journal of Financial Economics, 92, pp Anderson, D., Duru, A. and Reeb, D. (2010) Family preferences and investment policy: Evidence from R&D spending and capital expenditures, Working Paper, Temple University. Baek, J., Kang, J. and Park, K. (2004) Corporate governance and firm value: Evidence from the Korean financial crisis, Journal of Financial Economics, 71(2), pp Ball, R., Kothari, S. and Robin, A. (2000) The effect of international institutional factors on properties of accounting earnings, Journal of Accounting and Economics, 29(1), pp Ball, R., Robin, A. and Wu, J. S. (2003) Incentives versus standards: properties of accounting income in four East Asian countries, Journal of Accounting and Economics, 36, pp Beatty, A., Liao, S. and Weber, J. (2010) The effect of private information and monitoring on the role of accounting quality in investment decisions, Contemporary Accounting Research, 27 (1), pp Bianco, M., Golinelli, R. and Parigi, G. (2009). Family firms and investments, Working Paper, Bank of Italy and University of Bolonga. Biddle, G., Hilary, G. and Verdi, R. (2009) How does financial reporting quality improve investment efficiency? Journal of Accounting and Economics, 48 (2-3), pp Biddle, G. and Hilary, G. (2006) Accounting quality and firm-level capital investment, The Accounting Review, 81, pp Bjuggren, P., Dzansi, J. and Palmberg, J. (2008) Investment performance of Swedish listed family firms, Working Paper, Jönköpings International Business School. Bjuggren, P. O. and Palmberg, J. (2010) The impact of vote differentiation on investment performance in listed family firms, Family Business Review, 23(4), pp Bushman, R., Piotroski, J. and Smith, A. (2006) Capital allocation and timely accounting recognition of economic losses, Working Paper, University of North Carolina and University of Chicago. 23

25 Cao, Y., Myers, L. A. and Omer, T. C. (2012) Does company reputation matter for financial reporting quality? Evidence from restatements, Contemporary Accounting Research, forthcoming. Chen, S., Chen, X. and Cheng Q. (2008) Do family firms provide more or less voluntary disclosure? Journal of Accounting Research, 46(3), pp Chen, Y. R. and Ho, C. Y. (2009) The impact of family control and board characteristics on corporate policies, Journal of Management, 26 (1), pp Chen, F., Hope, O.-K., Li, Q. and Wang, X. (2011) Financial reporting quality and investment efficiency of private firms in emerging markets, The Accounting Review, 86(4), pp Carlock, R. S. (2010) When family businesses are best, Working Paper, INSEAD. Claessens, S., Djankov, S. and Lang, H. P. (2000) The separation of ownership and control in East Asian corporations, Journal of Financial Economics, 58, pp Claessens, S., Djankov, S., Fan, J. and Lang, H. P. (2002) Disentangling the incentive and entrenchment effects of large shareholdings, Journal of Finance, 57, pp Cucculelli, M. (2008) Owner identity and firm performance: Evidence from European companies, Working Paper, Università Politecnica delle Marche. DeAngelo, H. and DeAngelo, L. (2000) Controlling stockholders and the disciplinary role of corporate payout policy: A study of the times mirror company, Journal of Financial Economics, 56(2), pp Dechow, P. and Dichev, I. (2002) The quality of accruals and earnings: the role of accrual estimation errors, The Accounting Review, 77 (Supplement), pp Edmans, A. (2009) Blockholder trading, market efficiency, and managerial myopia, Journal of Finance, 64(6), pp Eklund, J. E., Palmberg, J. and Wiberg, D. (2011) Family ownership and returns on investment founders, heirs, and external managers, Working paper, Jönköpings International Business School. Faccio, M. and Lang, L. (2002) The ultimate ownership in western European corporations, Journal of Financial Economics, 65(3), pp Fama, E. and Jensen, M. (1985) Organizational forms and investment decisions, Journal of Financial Economics, 14, pp Fan, C. W., Tan, J., Guller, E., Garcia, B. and Ouek, A. (2011) Asian family businesses report, Credit Suisse. Fan, J. P. H. and Wong, T. J. (2002) Corporate ownership structure and the informativeness of accounting earnings in East Asia, Journal of Accounting and Economics, 33, pp Francis, J., LaFond, R. Olsson, P. and Schipper, K. (2005) The market pricing of accruals quality, Journal of Accounting and Economics, 39 (2), pp Francis, J. and Martin, X. (2010) Acquisition profitability and timely loss recognition, Journal of Accounting and Economics, 49(1-2), pp

26 Galeotti, M., Schiantarelli, F. and Jaramillo, F. (1994) Investment decisions and the role of debt, liquid assets and cash flow: Evidence from Italian panel data, Applied Financial Economics, 4, pp Gilson, R. and Gordon, J. (2003) Controlling shareholders, Working Paper, Columbia Law School, The Center for Law and Economic Studies, New York. Gugler, K. (2003) Corporate governance and investment, International Journal of the Economics of Business, 10, pp Gugler, K., Mueller, D. C. and Yurtoglu, B. B. (2007) Corporate governance and the determinants of investment, Journal of Institutional and Theoretical Economics, 163, pp James, H. (1999) Owner as manager, extended horizons and the family firm, International Journal of the Economics of Business, 6, pp Jensen, M. (1986) Agency costs of free cash flow, corporate finance, and takeovers, American Economic Review, 76 (2), pp Kedia, S. and T. Philippon (2009) The economics of fraudulent accounting, Review of Financial Studies, 22 (6), pp Klein, S. B. and Ward, J. L. (2012) Keeping the family in family firms: What sets family businesses apart, The Focus Online, Egon Zehnder International. Larcker, D., and Rosticus, T. O. (2010) On the use of instrumental variables in accounting research, Journal of Accounting and Economics, 49, pp Levine, R. (1997) Financial development and economic growth: views and agenda, Journal of Economic Literature, 35 (2), pp Li, K. K. and Tang, V. W. (2008) Earning quality and future capital investment: Evidence from discretionary accruals, Working Paper, Georgetown University. Lin, C. and Chang, C. (2009) Abnormal change of board members, family firms and fraud, The International Journal of Accounting Studies, 48, pp Lin, C., Wang, D. and Pan, C. (2012) The effect of free cash flow on investment decisions: From family firm and industry characteristics, Working Paper, National Taiwan University. McNichols, M. and Stubben, S. (2008) Does earnings management affect firms investment decisions?, The Accounting Review, 83(6), pp Morck, R. K. and Yeung, B. (2003) Agency problems in large family business groups, Entrepreneurship Theory and Practice, 27, pp Morgado, A. and Pindado, J. (2003) The underinvestment and overinvestment hypotheses: An analysis using panel data, European Financial Management, 9, pp Petersen, M. A. (2009) Estimating standard errors in finance panel data sets: Comparing approaches, The Review of Financial Studies, 22, pp Pindado, J., Requejo, I. and Torre, C. (2011) Family control and the investment-cash flow sensitivity: Empirical evidence from the Euro zone, Journal of Corporate Finance, 17(5), pp