Behavioral Corporate Finance － Models of

aaron

Published 2025-06-27 · 38 slides · 0 views

↓ Download

Behavioral Corporate Finance － Models of

Behavioral Corporate Finance － Models of - slide 1 of 38

Behavioral Corporate Finance － Models of - slide 2 of 38

Behavioral Corporate Finance － Models of - slide 3 of 38

Behavioral Corporate Finance － Models of - slide 4 of 38

Behavioral Corporate Finance － Models of - slide 5 of 38

Behavioral Corporate Finance － Models of - slide 6 of 38

Behavioral Corporate Finance － Models of - slide 7 of 38

Behavioral Corporate Finance － Models of - slide 8 of 38

Behavioral Corporate Finance － Models of - slide 9 of 38

Behavioral Corporate Finance － Models of - slide 10 of 38

Behavioral Corporate Finance － Models of - slide 11 of 38

Behavioral Corporate Finance － Models of - slide 12 of 38

Behavioral Corporate Finance － Models of - slide 13 of 38

Behavioral Corporate Finance － Models of - slide 14 of 38

Behavioral Corporate Finance － Models of - slide 15 of 38

Behavioral Corporate Finance － Models of - slide 16 of 38

Behavioral Corporate Finance － Models of - slide 17 of 38

Behavioral Corporate Finance － Models of - slide 18 of 38

Behavioral Corporate Finance － Models of - slide 19 of 38

Behavioral Corporate Finance － Models of - slide 20 of 38

Behavioral Corporate Finance － Models of - slide 21 of 38

Behavioral Corporate Finance － Models of - slide 22 of 38

Behavioral Corporate Finance － Models of - slide 23 of 38

Behavioral Corporate Finance － Models of - slide 24 of 38

Behavioral Corporate Finance － Models of - slide 25 of 38

Behavioral Corporate Finance － Models of - slide 26 of 38

Behavioral Corporate Finance － Models of - slide 27 of 38

Behavioral Corporate Finance － Models of - slide 28 of 38

Behavioral Corporate Finance － Models of - slide 29 of 38

Behavioral Corporate Finance － Models of - slide 30 of 38

Behavioral Corporate Finance － Models of - slide 31 of 38

Behavioral Corporate Finance － Models of - slide 32 of 38

Behavioral Corporate Finance － Models of - slide 33 of 38

Behavioral Corporate Finance － Models of - slide 34 of 38

Behavioral Corporate Finance － Models of - slide 35 of 38

Behavioral Corporate Finance － Models of - slide 36 of 38

Behavioral Corporate Finance － Models of - slide 37 of 38

Behavioral Corporate Finance － Models of - slide 38 of 38

Description

Behavioral Corporate Finance Models of Investors Irrationality 1 Outline Introduction DeAngelo et al. (2010) SEO proceeds, issuer characteristics, and estimated issuance probability Logit analysis The need for external finance Polk and

Download this presentation From Below

"Behavioral Corporate Finance － Models of" is the property of its rightful owner. Permission is granted to download and print the materials on this website for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.

Presentation Transcript

Behavioral Corporate Finance－ Models of Investor’s Irrationality 1

Outline Introduction
DeAngelo et al. (2010)
SEO proceeds, issuer characteristics, and estimated issuance probability
Logit analysis
The need for external finance
Polk and Sapienza (2009)
Investment and mispricing 2

Outline Discretionary accruals and investment
Cross-sectional tests
Efficient or inefficient investment? 3

Introduction An important strand of research on behavioral corporate finance asks whether irrational investors affect the financing and investment decisions of firms.
The financing decisions
1. Stein (1996) provides “market timing” view in issuing new equity: when a firm’s stock price is too high, the rational manager should issue more shares so as to take advantage of investor exuberance. Conversely, when the price is too low, he should repurchase shares. 4

Introduction 2. Empirical evidence
At the aggregate level. The share of new equity issues among total new issues is higher when the overall stock market is more highly valued. Baker and Wurgler (2000) show that the equity share is a reliable predictor of future stock returns: a high share predicts low, and sometimes negative stock returns.
At the individual firm level. B/M ratio of a firm is a good cross-sectional predictor of new equity issues. Baker and Wurgler (2002) show that firms with high 5

Introduction valuations issue more equity while those with low valuations repurchase their shares.
Long-term stock returns after an IPO or SEO are low, while long-term returns after a repurchase are high.
Baker and Wurgler (2002) show that a firm’s weighted-average historical market-to-book ratio is a good cross-sectional predictor of the fraction of equity in the firm’s capital structure.
The investment decisions
1.The critical question is whether the investor sentiment 6

Introduction affects actual investment decisions. There are several channels through which sentiment might affect investment.
Equity-dependent firms (Baker and Wurgler 2003)
Investor sentiment may distort investment for equity-dependent firms. When investors are excessively pessimistic, such firms may have to forgo attractive investment opportunities. As a result, it predicts that the investment of equity-dependent firms should be more sensitive to gyrations in stock price than the investment of non-equity dependent firms. 7

Introduction Managers might use investor exuberance as a cover for doing negative NPV “empire building” projects.
Managers who do not take the projects investors perceive as profitable may encounter the threat of takeover or being fired.
Managers put some weight on investors’ opinions.
2. Empirical evidence
Early studies produced little evidence of investment distortion.(Blanchard, Rhee and Summers 1993, Morck, Shleifer and Vishny 1993, Baker and Wurgler 2002a) 8

Introduction Polk and Sapienza (2001) find that the firms they identified as overvalued appear to investment more than other firms.
Baker, Stein and Wurgler (2003) test the cross-sectional prediction that equity-dependent firms will be more sensitive to stock gyrations. They find that equity-dependent firms have an investment sensitivity to stock prices about three times as high as that of non-equity dependent firms. 9

SEO proceeds and issuer characteristics – DeAngelo et al. (2010) DeAngelo et al. (2010) investigate the influences of market-timing opportunities and the corporate lifecycle stage on the probability that a firm conducts seasoned equity offering (SEO).
Data
This study analyses SEOs conducted by industrial firms for years 1973-2001. The samples include 4,291 SEOs. The main findings are based on samples with fewer than 4,291 samples as data availability conditions are imposed as necessary. 10

SEO proceeds and issuer characteristics – DeAngelo et al. (2010) 1. The size distribution of SEO proceeds (Table 1)
SEOs with the largest 10% of cash proceeds accounts for 48.6% of the total proceeds from all offerings. A large number of SEOs raise a small amount of cash and a modest number of SEOs raise a large amount of cash.
Current and former dividend payers account for 56.2% of SEOs in the top decile and 41.4% for all SEOs, contrary to the view that SEOs are mainly the province of young growth firms. 11

SEO proceeds and issuer characteristics – DeAngelo et al. (2010) 2. Univariate analysis of issuer and issue characteristics (Table 2)
Marketing-timing proxy variables
Standardized M/B ratio (raw M/B divided by median M/B for all firms), the prior 36(12)-month abnormal stock return, and the future 36-month abnormal stock return
Corporate lifecycle proxy variables. Dividend history and years listed. 12

SEO proceeds and issuer characteristics – DeAngelo et al. (2010) The data in panel A show that dividend groups are reasonable proxies for lifecycle stage (累積股利前500者:median listed year 42.3, M/B 0.98 v.s 從未發放股利者: median listed year 3, M/B 2.06).
The median issuer in the full sample is listed 5.1 years at the time of the SEO and has a standardized M/B of 1.71. Panel B documents 66.2% of SEOs are conducted by firms in the top two M/B quintiles. Panel C shows that 55% of the issuers are listed for less than five years and has a M/B ratio of 1.94. 13

SEO proceeds and issuer characteristics – DeAngelo et al. (2010) The long-horizon stock returns of the sample issuers also fit the general pattern documented in prior SEO studies. In panel D and E, firms that conducted SEOs tend to have experienced high abnormal stock returns over the most recent 36(12)-month period. In panel F, issuers tend to have low abnormal returns over the subsequent 36-month period. 14

Logit analysis Basic logit tests
The logit regressions use 27 years of data (1975-2001) on industrial firms’ SEO decisions. For a given firm in a given year, the dependent variable equals one if the firm conduct an SEO in that year, and zero if it does not.
Row A of Table 3 shows that the estimated SEO probability in a given year is significantly positively related to the firms’ standardized M/B and to its prior period market-adjusted return, and negatively related to its future market-adjusted return. 15

Logit analysis Row B to E of Table 3 present a variety of sensitivity checks of market timing opportunities. All conform the inferences based on row A.
Row F and G present the results of regressions that include years listed as an explanatory variable. As predicted by the lifecycle theory, the estimated probability of an SEO declines significantly with increases in the number of years listed.
Row H to J show the influences of dividend history (proxy for lifecycle stage). For top dividend payers, the 16

Logit analysis impacts of M/B and prior 36-month stock return become insignificant. The prior 12-month stock return and the future stock return is still significantly positively and negatively related to the estimated SEO probability, respectively.
The findings in row H to J support theories in which B/M influences stock issuance decisions because it proxies for growth opportunities while recent and future stock returns are more indicative of differential stock market-timing opportunities. 17

Logit analysis Relative impact of market timing versus stage of corporate lifecycle
Table 4 reports the estimated probability of an SEO as a function of hypothesized values of standardized M/B, prior and future stock returns, and the number of listed years (based on model G of table 3).
Row 1 considers a firm with neutral timing opportunities, while row 2-7 offer pairwise comparisons of extreme variation in each of the three market-timing variables. 18

Logit analysis The influence of timing opportunity alone (the far right column)
As future return decreases from 75% to -75%, the SEO probability increases by 1.5%; as prior return increases from -75% to 75%, the SEO probability increases by 1.9%; as M/B increases from 0.5 to 3, the SEO probability increases by 1.3%.
Row 11 and 12 compares the SEO probability under a highly favorable market-timing opportunity and a highly unfavorable opportunity. These estimates indicate that few firms with excellent opportunities conduct SEOs. 19

Logit analysis These estimates imply that market timing is, at best, an incomplete theory that requires modification to explain why so few firms conduct SEOs when they face attractive timing opportunities.
In relative terms, the SEO probability exceeds by 175.9% (5.1%/2.9%) when all market-timing variables indicate a favorable timing opportunity.
The influence of lifecycle stage
Row 1 reports that with neutral timing opportunity, for a firm listed one year SEO probability increases by 6.5% (in relative term 260% higher) than if it has been 20

Logit analysis listed 20 years.
The relative influence of market-timing and the number of year listed is best assessed by comparing row 11 and 12. A firm listed one year with highly unfavorable timing opportunities has a SEO probability of 6.5%, while a firm listed for 20 years with highly favorable timing opportunities has a SEO probability of 3.8% (increasing by 71.1%( 2.7%/3.8%)). As a result, the year-listed effect overrides the market-timing effect. 21

The need for external capital Basic intuition: If market timing were the primary motive for selling stock, only by chance would issuers be operating with seriously limited resources when a financing window opens, thus they would most often stockpile the cash proceeds from SEOs until suitable investment opportunities materialize.
Corporate cash balances in the years surrounding SEOs
1. Variables. Pro forma values of Cash and Total assets: the values these variables would take had the firm not received the SEO proceeds and all operating and other 22

The need for external capital financing decisions remains unchanged.
excess Cash/TA= Cash/TA – normal Cash/TA
2. Cash balances
Table 6 (row 1~3; 4~6) show that all sample partitions exhibit an SEO-induced increase and an immediate and almost-complete reversion in median cash ratios (7.2%~13.3%~8.8%; -0.1%~1.4%~0.0%).
To capture the pivotal importance of the SEO proceeds, this study calculates pro forma cash ratios in the year after SEO. The findings are in row 12~15. 23

The need for external capital All other decisions fixed, a majority of issuers would almost immediately run out of cash had they nor received the issue proceeds (median ratio in the year after the SEO is -4.2% or -11.7%). Without the SEO proceeds, 62.6% of the issuers would have negative Cash/TA, and 81.1% would have subnormal Cash/TA.
Abnormal change in cash = cash held-the cash the firm would have if it maintained its pre-SEO Cash/TA ratio. For each dollar raised in the SEO, the median issuer retains just 0.6% in excess cash in the year after the SEO (99.4% of the excess cash obtained 24

The need for external capital through the SEO is spent by the year after the SEO.)
This finding indicates that cash stockpile is the exception and not the rule in the sample, and it is consistent with the evidence in Rows 12~15 that most issuers face serious resource limitations.
The influence of capital expenditure
It is possible that issuers did not need the SEO proceeds, but rather they issued stock purely to time the market and quickly spent the proceeds on new investments that managers would not otherwise have undertaken. 25

The need for external capital Row 5 of Table 7 reports that the median issuer increases CapEx by 10% of total assets from the year before to the year after the SEO.
Panel B of Table 7 show that most issuers would experience an immediate cash shortfall even had they not increased capital expenditure following their SEOs (maintained the pre-SEO level of capital expenditures): 40.3% of issuers would run out of cash and 59.6% would have subnormal cash balances the year after the SEO. 26

Investment and Mispricing-Polk and Sapienza(2009) Following Stein (1996), this study offers a simple model to show how stock mispricing may have a direct effect on the investment policy of a firm.
Define shareholder’s expected utility at time 0 as eq.(1), the FOC of the manager’s problem (determining K at time 0) is as follows:

, α : the level of mispricing, p: mispricing disappears over time at the rate p, q: the arrival rate of the average shareholder 27

Investment and Mispricing-Polk and Sapienza(2009) The optimal investment level is when there is no mispricing which satisfies . When a firm is overpriced, the manager overinvests.
The incentive to overinvest increases as the expected duration of mispricing increases (p becomes smaller) and decreases as the horizon of the average shareholder lengthens (q becomes smaller). 28

Discretionary accruals and investment The basic specification regresses firm investment on discretionary accruals (the proxy for mispricing), a proxy for Tobin’s Q, and firm cash flow, controlling for firm and year fixed effect, eq.(3).
The proxy for mispricing exploits firms’ use of accrual accounting. Accruals represent the difference between a firm’s accounting earnings and its underlying cash flow. Several papers show a strong negative correlation between discretionary accruals and subsequent stock returns, suggesting that firms with high discretionary accruals are overpriced relative to otherwise similar firms. 29

Discretionary accruals and investment Accruals (ACCR) is measured by eq.(4). To capture the discretionary component of accruals, the authors follow Chan et al. (2001) such that

where accruals are scaled by total assets and model NORMALACCR as a constant proportion of firm sales.
Empirical Results
1. In Table 2, Panel A, column (1) displays the results of regression (3). When investment opportunities and cash flow are controlled for, firms with high discretionary accruals invest more. 30

Discretionary accruals and investment 2. Baker et al (2003) show that mispricing affects investment decisions through an equity channel. This study wants to test whether there is additional channel that links equity mispricing to investment.
In Table 2, Panel A, column (2), and all subsequent similar regressions, the authors control for cash from sale of common and preferred stocks scaled by Ki, t-1.
The discretionary accruals coefficient remains essentially the same as before, confirming that the catering channel has an independent effect. 31

Discretionary accruals and investment 3. The measure error problem in the proxy for investment opportunities ( Q )
First, include analysts’ consensus estimates of future earnings in the baseline regression. In column (3) through (5) of Table 2, Panel A, the authors add the ratio of consensus analyst forecast of cumulative firm profitability over assets one, two, and five years out to the baseline specification. In any case, discretionary accruals remain economically and statistically significant.
Second, measure Q at different time point. Qi,t controls 32

Discretionary accruals and investment for the change in Q over the investment period. Qt-2 and Qt-3 controls for the lagged effect of investment opportunities.
In Table 2, Panel A, column (6) through (8), the impact of discretionary accruals continue to be significant after controlling for the timing of Tobin’s Q variable. 33

Cross-sectional tests The model suggests that the greater the opacity of the firm (lower p) and the shorter the time horizon of the firm’s shareholders (larger q), the more likely managers are to cater investments.
1. The proxy for firm transparency: R&D intensity (higher intensity indicates less transparency)
The proxy for the time horizon of the firm’s shareholder: firm share turnover (higher turnover indicated shorter horizon)
2. In Table 3, Panel A, column (1) through (3) report results for all firms that have R&D data, for those firms 34

Cross-sectional tests below the median value of R&D intensity, and for those firms with R&D intensity above the median, respectively.
Firms that engage in a lot of R&D invest more when they have a lot of discretionary accruals (0.3052 vs. 0.1542).
3. In Table 3, Panel A, column (4) through (6) report results for all firms that have turnover data, for those firms with turnover below the yearly median, and for above-the-median firms, respectively.
The discretionary accruals coefficient is higher for high turnover firms (0.1726 vs.0.1154). 35

Efficient or inefficient investment In this model, because firm investment is linked to the market’s misvaluation of the firm’s equity, there is a negative relation between investment and subsequent risk-adjusted returns.
The authors estimate cross-sectional regressions of monthly stock returns on investment, Tobin’s Q, and a control for cash flow sensitivity, eq.(7). This regression ties return predictability to firm investment behavior.
Table 4, column (1), shows the result of estimating equation (7). Consistent with the model, firms that over invest (underinvest) on average have returns that are low (high). 36

Efficient or inefficient investment Table 4, column (2), includes three firm characteristics size, B/M, and momentum). These controls do not subsume the investment effect.
The model also predicts that this return predictability should be stronger for firms facing a greater degree of information asymmetry and/or having investors with shorter horizon.
In column (4), reestimate the relation by including an interaction variable between investment and an above-median R&D dummy variable.
In column (6), reestimate the relation by including an interaction variable between investment and the above- 37

Efficient or inefficient investment median share turnover dummy.
Both results confirm that the abnormal-investment effect in the cross-section of average returns is mainly in high R&D or high turnover firms. 38

Behavioral Corporate Finance － Models of

Related Topics

Share

Embed code

Download this presentation From Below

Presentation Transcript