Mutual Fund Trading Pressure:

Firm-Level Stock Price Impact and Timing of SEOs

Mozaffar Khan MIT Sloan School of Management

Leonid Kogan

MIT Sloan School of Management and NBER

George Serafeim Harvard Business School

March 2009

Abstract Mutual funds with large capital inflows exert upward price pressure on stocks they purchase as they invest their new capital. We use this price pressure as a relatively exogenous indicator of overvaluation to examine whether managers time SEOs and personal sales to take advantage of this mispricing. We document that: (i) inflow-driven buying pressure by mutual funds has a pronounced price impact on stocks; (ii) the odds of an SEO within the two quarters following inflow-driven mutual fund buying pressure increase by about 45% relative to the odds in quarters when there is no inflow-driven buying pressure; (iii) managers personally sell about 10% more stock in the two quarters following inflow-driven mutual fund buying pressure; and (iv) timed SEOs underperform other SEOs going forward. Collectively, the evidence suggests that managers are able to identify and exploit overvaluation resulting from mutual fund trading to transfer wealth from new to current shareholders. JEL Classification Code: G10, G11, G12, G14, G30, G32 Keywords: SEO, Market Timing, Price Impact, Trading Pressure, Mutual Fund

Khan is at 50 Memorial Dr., Cambridge, MA 02142; Ph. 617-252-1131; email mkhan@mit.edu. Kogan is at 50 Memorial Dr., Cambridge, MA 02142; Ph. 617-253-2289; email lkogan@mit.edu. Serafeim is at Soldier’s Field, Boston, MA 02163; email gserafeim@hbs.edu. We thank Jeff Callen, Hai Lu, Jeff Ng, Ricardo Reis, Sugata Roychowdhury, seminar participants at the London School of Economics, MIT and students in Khan’s doctoral seminar for helpful comments.

1. Introduction

Seasoned equity offerings have been widely studied in the literature, with little

emerging consensus on their determinants and economic consequences (Eckbo, Masulis

and Norli, 2007). Proposed determinants of SEOs include capital investments,1

refinancing,2 liquidity squeezes, corporate control,3 stock market microstructure4 and

timing by managers with private information that their stock is overvalued (Loughran and

Ritter, 1995, 1997; Graham and Harvey, 2001; Baker and Wurgler, 2002). In this paper

we propose a novel approach to testing the market-timing motive for SEOs, and provide

evidence of SEO timing. We also document the stock return performance of timed SEOs.

The main empirical challenge in tests of SEO timing is identifying overvalued

firms. Prior studies examining market timing have typically used high market-to-book

ratios or high past returns as identifiers of overvaluation. However, these studies

“continue to be hotly debated” (Baker, Ruback and Wurgler, 2007) because, as described

in detail in the next section, traditional indicators of overvaluation are correlated with

other determinants of SEOs. We identify a setting where overvaluation is relatively

exogenous to the firm. In particular, we identify overvalued stocks as those subject to

substantial buying pressure by mutual funds experiencing large capital inflows, but not

subject to widespread buying pressure by other mutual funds, and refer to these as stocks

subject to Inflow-driven Buying Pressure (IBP).

Mutual funds with large capital inflows are eager to invest the cash since

stockpiling cash makes it difficult for them to outperform their benchmarks (Coval and

1 For example, capital expenditures, acquisitions and restructurings. 2 For example, replacing existing securities or changing the capital structure. 3 For example, diluting voting rights. 4 For example, improving stock liquidity.

1

Stafford, 2007) and since they may be precluded by their investment mandate from

holding large cash balances. In addition, since funds follow specialized investment

strategies, they channel new funds into the stocks conforming to their strategies, rather

than diversifying their investments over the broader universe of stocks. We test the

hypothesis that, as these mutual funds invest their newly acquired cash, they exert

upward price pressure on the stocks they purchase. To distinguish overvaluation induced

by buying pressure from price appreciation driven by fundamental information about the

firms, we screen out stocks that are being widely purchased by all mutual funds and

retain only stocks subject to substantial buying pressure by mutual funds in the top decile

of capital inflows.

We find that Inflow-driven Buying Pressure (IBP) stocks experience steep

increases in cumulative abnormal returns during buying pressure quarters, followed by

steep declines of about 13% over the subsequent six quarters, consistent with

overvaluation due to inflow-driven buying pressure. This overvaluation can also be seen

as a persistent shift in the demand curve for the stock. In contrast, stocks subject to

widespread buying pressure by mutual funds other than those in the top decile of capital

flows (Non-IBP stocks) experience a cumulative decline in abnormal returns of less than

3% over the subsequent six quarters, consistent with widespread buying being driven

more by favorable firm-specific information than by demand shocks exogenous to the

firm.

If managers privately identify overvaluation and time SEOs to exploit the

overvaluation, we expect IBP-affected firms to exhibit a higher likelihood of SEOs

relative to other firms. We test this hypothesis using a Logit model of SEO choice that

2

controls for a number of determinants of SEOs, such as prior returns and the market-to-

book ratio. Our results show that the odds of an SEO within the two quarters (four

quarters) after the firm is affected by IBP increase by approximately 45% (78%) relative

to the odds in quarters with no IBP.

If IBP firms are overvalued we expect managers to time their own personal sales

as well. We test this using a model of insider sales, and find that managers sell

significantly more stock (about 10% more) in the two quarters following inflow-driven

buying pressure by mutual funds. This result aligns with and reinforces the inference

from the SEO tests.

We also examine the future stock return performance of timed SEOs, defined as

SEOs by IBP-affected firms, relative to all other SEOs. The future performance of

timed SEOs relative to other SEOs helps determine whether timing is a fortuitous

coincidence of demand for capital and the opportunity to raise cheap equity, or a case of

opportunity leading demand in the sense that managers take advantage of “windows of

opportunity” (Loughran and Ritter, 1995, 1997). In the latter case the new capital is

expected to be less productively employed because (a) there may be a delay in searching

for and fully exploiting productive opportunities, and (b) the marginal investment

undertaken by the firm because of temporary overvaluation of its equity is likely to have

a negative net present value. We note however that tests of future stock return

performance are not unambiguously tests of the ex ante inefficiency of investment

decisions.

We find that timed SEOs experience significantly negative three- and five- year

risk-adjusted returns. Their alphas are economically larger in magnitude (−0.73%

3

monthly at five year horizons) than the alphas for other SEOs (−0.29% monthly at five

year horizons), but the difference is not statistically significant. The lack of statistical

significance may be explained by the relatively small sample of SEOs subject to IBP, and

by noise in separating timers from non-timers. We identify timers as SEO firms subject

to Inflow-driven Buying Pressure by mutual funds in the recent two quarters, and non-

timers as all other SEOs. However, all other SEOs may also include timers responding to

alternative sources of overvaluation.

Our finding that timed SEOs underperform other SEOs is not induced

mechanically by the price pressure experienced by timed SEOs. Within the IBP sample,

we find that firms issuing SEOs underperform non-SEO firms on a risk-adjusted basis by

0.46% monthly at five year horizons. This is consistent with the most overvalued firms

among the IBP-affected firms having an SEO.

The stock return consequences of timed SEOs are economically important

because they suggest whether timing creates wealth or redistributes wealth. If timing

results in negative future abnormal stock returns, then a timed SEO leads to a wealth

transfer from new to current shareholders. Our finding of negative risk-adjusted stock

returns following timed SEOs supports the hypothesis that managers exploit windows of

opportunity presented by temporary overpricing of their firms’ equity to benefit current

shareholders at the expense of future shareholders.

The rest of this paper proceeds as follows. Section 2 briefly reviews the prior

literature on SEO timing and price pressure. Section 3 describes the identification of

overvalued and undervalued stocks as a result of flow-driven trading pressure by mutual

funds. Section 4.1 discusses tests of SEO timing, Section 4.2 discusses tests of the timing

4

of managers’ personal sales, and Section 4.3 documents post-SEO stock return

performance of timed SEOs. Section 5 describes supplementary analysis of the timing of

stock repurchases by firms and purchases by managers when there is outflow-driven

selling pressure by mutual funds, and also discusses robustness tests. Section 6

concludes with a summary, discussion of some implications of our findings and

suggestions for future research.

2. Related Literature

A number of papers have examined whether managers time the market in issuing

equity. The main empirical challenge is to identify mispriced stocks, and prior authors

have used ex ante and ex post methods to infer mispricing (Baker et al., 2007).

Ex ante methods include using a measure of fundamental value scaled by market

value such as the market-to-book ratio, or using prior returns, to identify overvalued

stocks. As emphasized in Baker et al. (2007), both measures are difficult to interpret. For

example, measuring fundamental value is difficult since accounting book values are

based on historical costs and subject to discretionary managerial accounting choices.

Further, the market-to-book ratio is correlated with many firm characteristics that may

drive financing policy, so high market-to-book ratios do not necessarily indicate

overvaluation that can be exploited by market timers. Prior returns as a measure of

mispricing face similar difficulties in interpretation. Firm with high prior returns may

have discovered valuable growth opportunities and harvesting these opportunities, rather

than market timing, drives the issuance decision.

5

Ex post methods rely on reversion in future abnormal returns to infer

overvaluation. For example, tests of long-horizon stock return performance following

SEOs find underperformance, suggesting that issuance occurred when the stock was

overpriced. One challenge to this interpretation is that risk changes may be associated

with SEOs. For example, Eckbo, Masulis and Norli (2000) suggest that equity issuance

lowers leverage and therefore systematic risk, leading to lower future returns.

In this paper we rely on the idea of price dislocation resulting from severe flow-

driven trading pressure by mutual funds in order to identify mispriced stocks using ex

ante information. This identification method yields a more reliable candidate for an

exogenous indicator of mispricing than the indicators discussed above, since overpricing

is driven by a shift in demand by mutual funds with large capital inflows rather than by

favorable fundamental information about the firm. Kraus and Stoll (1972), Harris and

Gurel (1986), Shleifer (1986) and Mitchell, Pulvino and Stafford (2004) provide

empirical evidence of price pressure, and Coval and Stafford (2007) in particular provide

evidence of persistent price pressure that is most directly related to ours. We discuss our

identification method in detail in the next section.

Consistent with our results, Frazzini and Lamont (2008) use mutual fund flows as

a measure of individual investor sentiment, and find that high sentiment predicts low

future returns. They also show that high sentiment is associated with increases in shares

outstanding in the next three years. Our paper differs in a number of ways: (a) our focus

is on the timing of SEOs and we establish a relation between SEOs and flow-driven price

pressure in a multiple regression setting where we control for the standard determinants

of SEOs; (b) we show that managers also time their personal trades to exploit flow-driven

6

price pressure; and (c) we show that managers respond to overvaluation through SEOs

and personal sales in a more timely manner, within two quarters of being affected by IBP.

Our paper is also related to Chen, Hanson, Hong and Stein (2007) who examine

whether hedge funds exploit mutual fund trading pressure through front-running. We

examine whether a different group of market participants, firm managers, exploit IBP

through SEOs and personal stock sales. Consistent with Chen et al. (2007), our results

suggest that “sophisticated” market participants are able to identify and exploit this

particular source of mispricing (price pressure due to inflow-driven mutual fund

purchases), although our data allows us to provide more direct evidence since we are able

to match SEOs and insider sales to IBP-affected stocks.

3. Mutual Fund Trading Pressure and Stock Price Impact

Mutual funds experiencing large capital inflows are eager to invest the new funds

quickly, since stockpiling cash may prevent them from tracking the performance of their

all-equity benchmarks, and their charter may require them to be fully invested in equities.

The buying pressure by such funds may create temporary price dislocation in prices of

the stocks they choose to buy. We follow the intuition in Coval and Stafford (2007) and

identify stocks that are subject to large net buying pressure by several mutual funds in the

top capital-flow decile as overvalued.

Specifically, for each stock held by mutual funds, we form a measure of trading

pressure as follows. First, we define mutual fund flows as {TAj,s – TAj,s-1(1+Rj,s-1)} /

TAj,s-1, where TA is total net assets of mutual fund j in month s, and R is the monthly

return for fund j in month s. We sum the monthly flows each quarter to obtain the

7

quarterly flow, flowj,t, of mutual fund j in quarter t. Monthly total net assets and returns

of mutual funds are obtained from CRSP, and quarterly mutual fund holdings are

obtained from Thomson Financial. Mutual funds are required to report their holdings

semi-annually, but approximately 60% of them report their holdings quarterly. To

calculate quarterly changes in holdings, we retain only contiguous fund-quarters in our

sample. In addition, we only consider open-ended U.S. equity funds and eliminate funds

with investment objective codes indicating bonds and preferred stocks, international

stocks, metals and municipal bonds from our sample.

Table 1 shows quarterly flows, expected flows, prior-year returns and the total

number of holdings for mutual funds, by flow decile, for 63,426 fund-quarters from 1990

to 2007. Expected flows are estimated by using eight quarters of lagged flows and lagged

fund returns as instruments, consistent with the prior literature (Ippolito, 1992; Chevalier

and Ellison, 1997; Sirri and Tufano, 1998; Coval and Stafford, 2007). Table 1 shows a

large spread in flows, ranging from 40.3% for the top decile to –17% for the bottom

decile. Expected flows decrease monotonically from 9.5% for the top decile to –4.2% for

the bottom decile, while prior-year fund returns decrease monotonically from 16.6% for

the top decile to 6.1% for the bottom decile. This confirms evidence in the prior

literature that (a) fund flows vary monotonically with past fund performance, and (b) the

flow-performance relation is asymmetric in that inflows due to good past performance are

much larger in magnitude than outflows due to poor past performance (Ippolito, 1992;

Chevalier and Ellison, 1997; Sirri and Tufano, 1998). Finally, funds with extreme flows

have fewer holdings on average than funds in the middle deciles of flows, suggesting that

8

for some funds extreme performance may be associated with risk from their more

concentrated positions.

We calculate trading pressure for stock i in quarter t as:

Pressurei,t =

∑j

,0(max( ∆holdingj,i,t) | flowj,t>90th percentile) – ∑j

,0(max( −∆holdingj,i,t) | flowj,t<10th percentile)

Shares Outstandingi,t-1

where i indexes the stock, j indexes the mutual fund and t indexes the calendar quarter.

This measure is similar to those used in Coval and Stafford (2007) and Chen, Hanson,

Hong and Stein (2007). Intuitively, trading pressure is interpreted as buying pressure

when funds with large inflows are net buyers of the stock, and as selling pressure when

funds with large outflows are net sellers of the stock. In a sense, Pressure is a measure of

excess demand from mutual funds with large capital flows.

To distinguish flow-motivated trading from information-motivated trading, we

calculate unforced pressure for stock i in quarter t as:

UPressurei,t = { ∆holding∑j

j,i,t | 10th percentile≤flowj,t≤90th percentile } / Shares Outstandingi,t-1

This variable captures net trading activity in a stock by mutual funds in the middle eight

deciles of flows, or widespread net trading. Information-driven purchases are identified

as stocks in the top decile of UPressure. UPressure is similar to measures used in

Lakonishok, Shleifer and Vishny (1992) and Wermers (1999) to identify mutual fund

demand imbalances.

We sort stock-quarters into deciles of Pressure and UPressure, and identify IBP-

affected stocks as those in the top decile of Pressure but in the middle three deciles

9

(deciles four, five, and six) of UPressure. In other words, IBP-affected stocks are those

that are subject to large buying pressure by mutual funds with extreme inflows, but that

are not subject to widespread net trading pressure by other mutual funds.

Table 2 shows quarterly mean abnormal returns from quarters t−4 to t+6 for

stocks subject to mutual fund buying pressure in quarter t=0 (Pressure is calculated in

quarter t=0). In Panel A, abnormal returns are defined as returns in excess of the value-

weighted average return on all stocks held by all mutual funds that quarter. In Panel B,

abnormal returns are industry-adjusted returns. The table shows results for 1733 stock-

quarters with IBP-affected stocks (Inflow-driven Pressure), and 12,742 stock-quarters

with stocks subject to information-driven buying pressure (Non-Inflow-driven Pressure).

The sample in Table 2 covers the period from 1990 through 2002, since we wish to test

the 3- and 5- year future performance of SEOs in response to price pressure, but results

are robust to post-2002 observations. We calculate mean abnormal returns each quarter

for the portfolio of Inflow-driven Pressure and Non-Inflow-driven Pressure stocks, and

use the time series of portfolio abnormal returns for statistical inference to control for

cross-sectional correlation (e.g., Fama and Macbeth, 1973; Coval and Stafford, 2007).5

Both Panels A and B of Table 2 show that stocks subject to IBP experience

significantly positive and large abnormal returns in buying quarters followed by

significantly negative and large abnormal returns in subsequent quarters as the buying

pressure subsides. In Panel A, IBP-affected stocks have cumulative abnormal returns of

-12.77% (p-value<0.01) while in Panel B they have cumulative industry-adjusted returns

of -13.46% (p-value<0.01), in quarters t+1 to t+6. In contrast, stocks subject to Non-

5 Results are robust to using the pooled cross-sectional and time-series panel to calculate the mean for each event quarter, with standard errors clustered by calendar time to control for cross-sectional correlation.

10

Inflow-driven Pressure experience small negative abnormal returns after quarter t=0, with

cumulative abnormal returns from t+1 to t+6 of -2.65% (p-value<0.10) in Panel A and

-2.88% (p-value<0.10) in Panel B. These abnormal return patterns are consistent with

inflow-driven buying pressure inducing overvaluation, and widespread mutual fund

buying being driven by favorable firm-specific information that is quickly impounded

into market prices.

Figures 1 and 2 show cumulative average abnormal returns (CAAR) for Inflow-

driven Pressure and Non-Inflow-driven Pressure stocks. In Figure 1, abnormal returns

are returns in excess of the returns of all stocks held by all mutual funds that quarter,

while in Figure 2 abnormal returns are industry-adjusted returns. We sum average

abnormal returns over consecutive quarters to obtain the cumulative average abnormal

returns shown in Figures 1 and 2. The CAAR patterns in Figures 1 and 2 suggest that

flow-driven mutual fund purchases induce substantial overvaluation, or a shift in the

demand curve for a stock. In contrast, widespread buying by mutual funds (non-inflow-

driven purchases) does not seem to result in overvaluation.6

4. Main Tests

In this section we present results of tests of SEO timing, timing of insider sales,

and post-SEO future performance of timed SEOs.

4.1 Timing of SEOs

We test whether firms identify and exploit the overvaluation induced by mutual

fund flow-driven buying pressure by issuing seasoned equity within two quarters of the 6 Figures 1 and 2 depict firm-level, not mutual fund-level, performance.

11

stock being subject to buying pressure by mutual funds with large capital inflows but not

subject to widespread buying by other mutual funds (SEO occurs in t and buying pressure

in t-2 or t-1). We do not examine SEOs contemporaneous with buying pressure to avoid

confounding inferences through any reverse causality (i.e., that mutual funds may be

buying because firms are issuing equity, rather than the reverse, which is our hypothesis).

We test our hypothesis of SEO timing by estimating a Logit regression of the issuance

decision in quarter t, including an indicator dummy for stock-quarters with inflow-driven

buying pressure (but not widespread buying pressure) in quarters t-1 or t-2.

Following the prior literature (e.g., Eckbo et al., 2000), we obtain SEO data from

the SDC database. After retaining only common stock issuances that trade on the NYSE,

AMEX or NASDAQ, and excluding secondary offerings in which no new shares are

issued, investment trusts (e.g., REIT’s), American Depositary Receipts and utilities (SIC

codes 4910-4939) we are left with 3912 SEOs between 1990 and 2002. We exclude

utilities since the duration of the regulatory approval process limits the ability of utilities

to time SEOs in response to temporary overpricing (Eckbo and Masulis, 1992).

Accounting data are obtained from Compustat Fundamentals Quarterly. We

exclude all utilities and securities that do not have a share code of 10 or 11 in CRSP.

Insider trading data are obtained from the Thomson Financial Insider database. The final

sample includes 239,112 firm-quarters from 1990 through 2002, and 155,624 firm-

quarters when insider trading data is used.

Table 3 shows descriptive statistics for the full sample (Panel A), for IBP-affected

stocks (Panel B) and for stocks subject to widespread buying pressure by mutual funds

other than those in the top decile of capital flows (Panel C). In Panel A, Return on assets

12

(ROA) is defined as operating income before depreciation over total assets at the end of

the same quarter in the prior year, and has a mean of 1.8%. One-year price appreciation

(1-year Price appreciation) is defined as the growth in stock price over a one-year period

ending at the end of the current quarter, and has a mean of 14.7%.7 Book-to-market

(BTM) is book value of shareholders’ equity over market value of equity at the end of the

same quarter in the prior year, and has a mean of 0.699. Size is the natural logarithm of

total assets at the end of the same quarter in the prior year, and has a mean of 5.17.

Leverage is long-term debt plus long-term debt in current liabilities, over total assets, at

the end of the same quarter in the prior year, and has a mean of 0.21.

Also in Panel A of Table 3, Dividend yield is the quarterly dividend per share

divided by the stock price in the same quarter last year, and has a mean of 0.004. The

median dividend yield is zero, suggesting that most firms do not pay dividends, which is

consistent with the evidence in Skinner (2008). Cash is cash and short-term investments

over total assets at the end of the same quarter in the prior year, and has a mean of 0.158.

InsidSale (All) is the ratio of shares sold by all insiders at firm i in quarter t to the sum of

shares purchased and shares sold by all insiders in that firm-quarter, and has a mean of

0.382. InsidSale (Execs) is the ratio of shares sold by the top 5 executives at firm i in

quarter t to the sum of shares purchased and shares sold by the top 5 executives in that

firm-quarter, and has a mean of 0.367. InsidHolding is the ratio of shares held by all

insiders to the total number of shares outstanding, and has a mean of 0.027.

Panels B and C show that IBP-affected stocks and stocks subject to widespread

buying are similar in all characteristics in Table 3. The only pronounced difference is in

7 Our results remain virtually unchanged when we use 1-year stock returns instead of 1-year price appreciation.

13

1-year price appreciation, which is 53% for IBP-affected stocks and 28.7% for stocks

subject to widespread buying. This suggests that the abnormal return patterns for these

two groups described in the previous section are due not to differences in fundamental

information about these stocks, but rather, to overvaluation of IBP stocks due to flow-

induced mutual fund buying pressure.

Table 4 shows the results of the Logit regression for the seasoned equity issuance

choice. The dependent variable is 1 if the firm has a seasoned equity offering (SEO) in

quarter t, and zero otherwise. Buy Pressure (t−1, t−2), our main independent variable of

interest, is a dummy that equals 1 if the stock was subject to IBP in quarter t−1 or quarter

t−2. The other independent variables are determinants of the equity issuance decision

suggested in the literature, and are described further below. The regression includes

industry, year and quarter fixed effects, and standard errors are clustered on both firm and

time (two-dimensional clustering) to control for cross-sectional and time-series

correlation (Petersen, 2009).

The main result in Table 4 is that Buy Pressure (t−1, t−2) is positive with a p-

value of less than 1% in a one-tailed significance test. This suggests that, ceteris paribus,

firms are significantly more likely to have SEOs in the two quarters following flow-

driven buying pressure. The associated odds ratio is 1.454, meaning that the odds of an

SEO in the two quarters following buying pressure are 45.4% higher than the odds of an

SEO in other quarters, which is an economically significant magnitude. This result holds

after controlling for prior price appreciation and the book-to-market ratio (BTM). The

interpretation is as follows. Consider two firms with similar prior price appreciation and

BTM, only one of which was IBP-affected. Under our hypothesis the IBP-affected firm

14

is overvalued, while, as described earlier, high price appreciation and low BTM may be

interpreted as due to fundamental news. The timing hypothesis suggests Buy Pressure

(t−1, t−2) should load after controlling for fundamental news in prior price appreciation

and BTM, and the hypothesis is supported in Table 4.

Also in Table 4, the probability of issuing equity is significantly increasing in the

one-year stock price appreciation (1 year price appreciation), and significantly decreasing

in the book-to-market ratio (BTM), consistent with firms issuing equity after they have

experienced high returns (e.g., Asquith and Mullins, 1986; Loughran and Ritter, 1995) or

when their market values are high relative to book values (e.g., Jenter, 2005; DeAngelo,

DeAngelo and Stulz, 2007). Larger firms are more likely to issue equity, which is

consistent with larger firms facing lower average issuance costs. Leverage loads

significantly positively, consistent with “financially constrained” firms being more

“equity-dependent” (e.g., Baker, Stein and Wurgler, 2003). The loading on the Dividend

yield is significantly negative and Cash loads significantly positively, consistent with

equity-issuing firms retaining earnings and accumulating cash in addition to raising

external equity ahead of capital investments. ROA is insignificant. Overall, Table 4

provides support for the hypothesis that firms time equity issuances to exploit exogenous

overvaluation.

4.2 Timing Personal Stock Sales by Managers

If managers exploit price pressure by timing SEOs, we expect them to time their

personal sales similarly. We test whether managers sell more of their personal shares

within two quarters of the quarter in which their stock is subject to IBP. The hypothesis

is tested by estimating regressions of insider sales in quarter t, including an indicator

15

dummy for stock-quarters with IBP in quarters t−1 or t−2.

One potential limitation on the power of our test to reject the null of no timing by

managers is that the ability of insiders to sell their shares is frequently restricted by firms

to certain short windows, e.g., within three to twelve days after quarterly earnings

announcements (Bettis, Coles and Lemmon, 2000), or is restricted by the requirement to

maintain a certain minimum level of holdings. We expect this is unlikely to be an issue

because: (a) we examine insider sales over the two quarters following flow-driven buying

pressure, not within a narrow window of a few days; and (b) this limitation biases against

our ability to reject the null.

Table 5 shows results of a panel regression with InsidSale, the ratio of shares sold

to the sum of shares sold and purchased, as the dependent variable.8 The second and

third columns report results for all insiders, while the last two columns report results for

the top 5 executives. Since the results are similar, we discuss the results for all insiders

only. All trades are open market trades. The regression includes year-, quarter- and

industry-fixed effects, and standard errors are clustered on both firm and time (two-

dimensional clustering). The main result in Table 5 is that Buy Pressure (t−1, t−2) is

significantly positive at less than 1% one-tailed, with a coefficient of 0.0372. Since the

mean of InsidSale (All) in Table 3 is 0.382, this suggests that insider sales increase by

0.0372/0.382 = 10% in the two quarters following IBP, which is an economically

significant magnitude.

Also in Table 5, insider sales are increasing in firm size consistent with Seyhun

(1986), Rozeff and Zaman (1988) and Core et al. (2006). InsidSale is significantly

8 Our dependent variable, the sales ratio, is similar to that in Rozeff and Zaman (1998) and Piotroski and Roulstone (2005) who use the purchase ratio, calculated as shares purchased divided by the sum of shares purchased and sold.

16

positively related to the past one-year price appreciation (1-year Price appreciation),

suggesting that insiders sell relatively more than they purchase when the stock price is

high relative to the past. InsidSale (All) is nearly monotonically decreasing with the

BTM decile which is consistent with the prior literature (e.g., Rozeff and Zaman, 1998;

Piotroski and Roulstone, 2005; Jenter, 2005). The BTM decile dummies are labeled

BTM1, BTM2 and so on, with BTM1 being low and BTM10 being high book-to-market.

BTM10 is omitted from the regression and so it is the reference or base decile. This

result suggests insiders at high growth firms sell relatively more than they purchase

compared to insiders at value firms, consistent with exit or diversification needs of

insiders at high growth firms.

As in Core et al. (2006) we control for insiders’ normal propensity to trade with

lagged InsidSale, and find that InsidSale (t-4) is significantly positive at less than 1% in

Table 5. InsidSale (t-4) effectively controls for unidentified omitted variables in our

regression. Finally, we also control for the level of insider holdings, InsidHolding, and

find that it loads significantly positively at less than 1%, suggesting insiders sell more

when they have higher exposure to firm-specific risk through higher holdings.

Overall the results in Table 5 align with those reported in Table 4 and suggest

insiders are able to identify overvaluation due to inflow-driven buying pressure by mutual

funds and subsequently to time both firm-level equity issuances and personal stock sales.

4.3 Future Return Performance of Timed SEOs

Following the extensive prior literature on the underperformance of SEOs, we

document post-SEO performance of timed SEOs, defined as SEOs by IBP-affected firms.

If firms exploit windows of opportunity to time equity issuances ahead of the demand for

17

capital, we expect that such firms are more likely to invest in low or negative NPV

projects and to have low future abnormal returns.

We measure abnormal returns as the intercept (or Jensen’s alpha) from calendar-

time Fama and French (1993) factor model regressions. Each quarter we form a portfolio

of all timers, and non-timers, with an SEO in the last three or the last five years. An SEO

enters the portfolio for the first time in the quarter following the SEO. The second and

third columns of Table 6 (Timed SEOs and Other SEOs) show the results of regressing

monthly timer and non-timer portfolio excess returns on an intercept and the three Fama

and French (1993) factors. Panel A shows results for three year horizons while Panel B

shows results for five year horizons.

In Table 6, Timed SEO alphas are significantly negative. While this is consistent

with the documented general underperformance of SEOs (e.g., Loughran and Ritter,

1995; Spiess and Affleck-Graves, 1995), risk-adjusted returns on timed SEOs are

particularly low, providing further evidence that IBP-affected firms issue equity to take

advantage of overvaluation. At three-year horizons timers have a monthly alpha of -

0.82% (p-value<0.1) compared to -0.57% (p-value<0.01) for all other SEOs, while at

five-year horizons timers have a monthly alpha of -0.73% (p-value<0.05) compared to

-0.29% (p-value<0.05) for all other SEOs. The SEO underperformance magnitudes are

consistent with the prior literature (e.g., Ritter, 2003).

The difference in performance of timed and other SEOs is economically

significant but not statistically significant. Our ability to differentiate future return

performance of timed SEOs from that of all other SEOs is limited by noise in our

separation of timers from non-timers. Specifically, we define timers as those that have an

18

SEO in the two quarters following IBP. However, other SEOs may include timers not

covered by our definition.

The fourth column of Panels A and B of Table 6 reports alphas for firms in the

IBP-affected sample that did not have an SEO (Non-SEO Buy-Pressured). Non-SEO

IBP-affected stocks have monthly alphas of -0.34% (p-value<0.1) at three year horizons

and -0.27% (p-value<0.1) at five year horizons. This difference in performance provides

further evidence that the relatively more overvalued among the IBP-affected firms are

more likely to issue an SEO.

Overall, the results in Table 6 support the hypothesis that timed SEOs

opportunistically redistribute wealth from new to current shareholders.

5. Additional Analysis

In this section we test whether managers also identify and exploit undervaluation

to time corporate and personal share purchases. We also discuss robustness tests.

5.1. Timing Firm Repurchases and Insider Purchases

While the focus of our paper is on timing of SEOs, we also examine whether

managers time firm repurchases (Brav, Graham, Harvey and Michaely, 2005) and their

own personal purchases to exploit adverse price pressure due to selling pressure by

mutual funds with extreme capital outflows.

Following Fama and French (2001) and Skinner (2008), net firm repurchases are

calculated from the change in common Treasury stock. An increase in common Treasury

stock is a net repurchase. If Treasury stock is zero in the current and prior quarters, we

use the difference between stock purchases and stock issuances obtained from the

19

Statement of cash Flows. We use a repurchase dummy equal to 1 for a net repurchase

and zero otherwise, and estimate a Logit regression of the repurchase decision with the

same independent variables as used in our SEO Logit regression. Unreported tests show

the odds of firms repurchasing shares significantly increase by about 20% in the two

quarters following outflow-driven selling pressure compared to the odds in quarters

without outflow-driven selling pressure. The smaller odds ratio, compared to the odds

ratio for timed SEOs reported earlier, suggests firms have more flexibility in the decision

to raise capital than in the decision to distribute capital. Another interpretation of this

result is that the marginal effect of mispricing on the repurchase decision is expected to

be lower since firm repurchases are more common than SEOs.

The insider purchase regression is similar to the insider sale regression, but in this

case we do not find evidence that managers buy significantly more shares following

outflow-driven selling pressure. A Sale Pressure (t−1, t−2) dummy that equals 1 if the

stock was subject in quarter t−1 or quarter t−2 to severe selling pressure by mutual funds

in the lowest decile of capital flows (flow-driven selling pressure) is statistically

insignificant. Compared to the significant insider sale result reported in Section 4.2, one

interpretation of this result is that insiders perceive possible over-exposure as more costly

than possible under-exposure to firm-specific risk.

Overall these results are consistent with the inference that flow-driven trading by

mutual funds exerts price pressure on stocks and that managers are able to identify and

exploit this price pressure.

5.2. Robustness Tests

Our results are robust to a number of variations in variable measurement. First,

20

results are intact when we calculate Pressure and UPressure using average lagged

quarterly trading volume (over the prior two quarters) as the denominator. Second, we

obtain robust results when we define insider sales as net sales (shares sold minus

purchased) scaled by shares outstanding. Third, results are robust when we expand the

managerial response window to four quarters (quarters t+1 to t+4) following the IBP

quarter. In this case we examine SEOs and insider sales in quarters t+1 to t+4.

6. Conclusion

We find that stocks subject to buying pressure by mutual funds experiencing large

capital inflows, but not widespread buying pressure by other mutual funds, experience

substantial upward price impact. We use widespread mutual fund buying pressure as an

indicator of informed trading. Therefore, the result suggests that stock prices change in

response to uninformed shifts in demand.

The price effects of mutual fund buying pressure are sufficiently long-lived to

allow managers who are able to identify the overvaluation to time SEOs and personal

share sales. Thus, inflow-driven mutual fund buying pressure is a useful overvaluation

identifier for SEO timing studies because it is relatively exogenous to the firm. We find

the odds of an SEO in the two quarters following the occurrence of flow-driven buying

pressure are 45% higher than the odds in quarters without flow-driven buying pressure,

and managers personally sell about 10% more shares during the same period. This

suggests that managers are able to identify and exploit inflow-driven buying pressure to

time SEOs and personal share sales.

21

In addition, timed SEOs have negative risk-adjusted returns, lower than that of all

other SEOs and of firms affected by buying pressure that do not issue an SEO. This is

consistent with managers exploiting temporary overvaluation by issuing an SEO to

transfer wealth from new to current shareholders (Loughran and Ritter, 1995, 1997).

Our evidence of long-lived price impact of uninformed demand shifts, while

consistent with empirical evidence in Coval and Stafford (2007) and the arguments of

Shleifer (1986), is intriguing. One possible conclusion is that arbitrage was unsuccessful

at flattening the demand curve. Reasons for limits to arbitrage include the unavailability

of close substitutes, specialization among arbitrageurs combined with a limited number of

arbitrageurs, lingering differences in investor opinion about the true value of the stock

and short sales constraints. Alternatively, as Coval and Stafford (2007) point out, market

participants may have been unaware of the return pattern induced by mutual fund flows

because the relevant data was not available at that time. As we show, despite the data

limitations, some firm managers were able to identify overvaluation in the equity of their

own firm and react to it by issuing additional shares or selling shares from their personal

account. The persistence of the price impact presents one opportunity for future research.

Our findings have a number of additional implications for future research. First,

in an international context, smaller markets with fewer close substitutes for individual

stocks may experience greater price impact from uninformed shifts in demand, which

suggests that the patterns of predictability we identify may be even stronger in such

markets. Second, the possibility of persistent stock price dislocation due to uninformed

demand shifts has implications for managerial performance evaluation and the optimal

sensitivity of managerial compensation to short-run stock returns. Third, from a

22

corporate governance perspective, our findings suggest managers are sometimes better

informed than the market and are able to identify market misvaluations, which has

implications for the level of discretion they are allowed in responding to price

movements through a variety of corporate decisions. Other research opportunities

include examining the bond price impact of mutual fund trading pressure and the

likelihood of subsequent debt issuances, and examining the effect of mutual fund trading

pressure on the use of cash versus stock in corporate acquisitions.

23

References Asquith, Paul and David Mullins. 1986. Equity issues and offering dilution. Journal of Financial Economics, 15: 61-89. Baker, Malcolm and Jeffrey Wurgler. 2002. Market timing and capital structure. Journal of Finance, 57 (1, Feb.): 1-32. Baker, Malcolm, Richard Ruback and Jeffrey Wurgler. 2007. Behavioral corporate finance: a survey. Forthcoming, Handbook of Corporate Finance: Empirical Corporate Finance: North Holland/Elsevier. Baker, Malcolm, Jeremy Stein and Jeffrey Wurgler. 2003. When does the market matter? Stock prices and the investment of equity-dependent firms. Quarterly Journal of Economics, (Aug): 969-1005. Bettis, J.C., J.L. Coles and M.L. Lemmon. 2000. Corporate policies restricting trading by insiders. Journal of Financial Economics, 57: 191-220. Brav, Alon, John Graham, Campbell Harvey and Roni Michaely. 2005. Payout policy in the 21st century. Journal of Financial Economics 77:483-527. Chen, Joseph, Samuel Hanson, Harrison Hong and Jeremy Stein. 2007. Do hedge funds profit from mutual-fund distress? Working paper, USC, Harvard and Princeton. Chevalier, Judith and Glenn Ellison. 1997. Risk taking by mutual funds as a response to incentives. Journal of Political Economy 105 (6, Dec.): 1167-1200. Core, John, Wayne Guay, Scott Richardson and Rodrigo Verdi. 2006. Stock market anomalies: what can we learn from repurchases and insider trading? Review of Accounting Studies 11 (1): 49-70. Coval, Joshua and Erik Stafford. 2007. Asset firesales (and purchases) in equity markets. Journal of Financial Economics 86:479-512. DeAngelo, Harry, Linda DeAngelo and Rene Stulz. 2007. Fundamentals, market timing and seasoned equity offerings. Working paper, University of Southern California and Ohio State University. Eckbo, B. Espen and Ronald Masulis. 1992. Adverse selection and the rights offer paradox. Journal of Financial Economics, 32: 293-322. Eckbo, B. Espen, Ronald Masulis and Oyvind Norli. 2000. Seasoned public offerings: resolution of the new issues puzzle. Journal of Financial Economics, 56: 251-292.

24

Eckbo, B. Espen, Ronald Masulis and Oyvind Norli. 2007. Security offerings. Forthcoming, Handbook of Corporate Finance: Empirical Corporate Finance, Vol. 1: North Holland/Elsevier. Fama, Eugene and James Macbeth. 1973. Risk, return, and equilibrium: empirical tests. Journal of Political Economy, 81: 607-636. Fama, Eugene and Kenneth French. 1993. Common risk factors in the returns on stocks and bonds. Journal of Financial Economics 33, 3-56. Fama, Eugene and Kenneth French. 2001. Disappearing dividends: changing firm characteristics or lower propensity to pay? Journal of Financial Economics 60:3-43. Frazzini, Andrea and Owen Lamont. 2008. Dumb money: mutual fund flows and the cross-section of stock returns. Journal of Financial Economics, 88(2): 299-322. Graham, John and Campbell Harvey. 2001. The theory and practice of corporate finance: evidence from the field. Journal of Financial Economics 60:187-243. Harris, Lawrence and Eitan Gurel. 1986. Price and volume effects associated with changes in the S&P 500 list: new evidence for the existence of price pressures. Journal of Finance 41 (4, Sep.): 815-829. Ippolito, Richard. 1992. Consumer reaction to measures of poor quality: evidence from the mutual fund industry. Journal of Law and Economics 35 (1, Apr.):45-70. Jenter, Dirk. 2005. Market timing and managerial portfolio decisions. Journal of Finance, 60 (4): 1903-1949. Kraus, Alan and Hans Stoll. 1972. Price impacts of block trading on the New York Stock Exchange. Journal of Finance 27 (3, June):569-588. Lakonishok, Josef, Andrei Shleifer and Robert Vishny. 1992. The impact of institutional trading on stock prices. Journal of Financial Economics, 32: 23-44. Loughran, Tim and Jay Ritter. 1995. The new issues puzzle. Journal of Finance, 50 (1, Mar.): 23-51. Loughran, Tim and Jay Ritter. 1997. The operating performance of firms conducting seasoned equity offerings. Journal of Finance, 52 (5, Dec.): 1823-1850. Mitchell, Mark, Todd Pulvino and Erik Stafford. 2004. Price pressure around mergers. Journal of Finance, 59: 31-63. Petersen, Mitchell. 2009. Estimating standard errors in finance panel data sets: comparing approaches. Review of Financial Studies, 22: 435-480.

25

Piotroski, Joseph and Darren Roulstone. 2005. Do insider trades reflect both contrarian beliefs and superior knowledge about future cash flow realizations? Journal of Accounting and Economics, 39(1): 55-82. Ritter, Jay. 2003. Investment banking and securities issuance, in Handbook of the Economics of Finance, vol. 1A, Constantinides, Harris and Stulz, eds. Rozeff, Michael and Mir Zaman. 1988. Market efficiency and insider trading: new evidence. Journal of Business, 61 (1): 25-44. Rozeff, Michael and Mir Zaman. 1998. Overreaction and insider trading: evidence from growth and value portfolios. Journal of Finance, 53: 701-716. Seyhun, H. Nejat. 1986. Insider profits, costs of trading, and market efficiency. Journal of Financial Economics, 16 (6): 189-212. Shleifer, Andrei. 1986. Do demand curves for stocks slope down? Journal of Finance, 41: 579-590. Shleifer, A. and R. Vishny. 1992. Liquidation values and debt capacity: a market equilibrium approach. Journal of Finance 47: 1343—1366. Sirri, Erik and Peter Tufano. 1998. Costly search and mutual fund flows. Journal of Finance 53 (5, Oct.):1589-1622. Skinner, Douglas. 2008. The evolving relation between earnings, dividends and stock repurchases. Journal of Financial Economics 87:582-609. Spiess, D. Katherine and John Affleck-Graves. 1995. Underperformance in long-run stcok returns following seasoned equity offerings. Journal of Financial Economics, 38: 243-267. Wermers, Russ. 1999. Mutual fund herding and the impact on stock prices. Journal of Finance, 59(2): 581-622.

26

Table 1: Mutual Fund Flow Predictability and Prior Performance Mutual funds are sorted quarterly into deciles of capital flows. The sample consists of 63,426 fund-quarters from 1990 to 2007. Flow is calculated as as {TAj,s – TAj,s-1(1+Rj,s-1)} / TAj,s-1, where TA is total net assets of mutual fund j in month s, and R is the quarterly return for fund j in month s. Monthly flows are summed to obtain the quarterly flow, flowj,t, of mutual fund j in quarter t. E[Flow] is the expected quarterly flow from a pooled regression model with eight lags of quarterly flows and quarterly returns as predictors. Prior Fund Return is the fund return in the last year. Avg # Holdings is the average number of stocks in a fund quarter. Prior Fund Avg # Flow Decile Flow% E[Flow]% Return (%) HoldingsInflow 40.3% 9.5% 16.6% 106.9 9 14.1% 7.3% 14.6% 122.1 8 7.0% 4.8% 13.2% 148.7 7 3.3% 2.4% 12.1% 144.8 6 1.0% 0.7% 11.0% 144.1 5 -0.7% -0.6% 9.8% 126.0 4 -2.3% -1.8% 9.1% 117.1 3 -4.1% -2.8% 7.8% 106.5 2 -6.8% -3.5% 6.3% 98.7 Outflow -17.0% -4.2% 6.1% 103.4

27

Table 2: Quarterly Abnormal Stock Returns due to Mutual Fund Buying Pressure The table shows mean quarterly abnormal returns from quarters t-4 to t+6 for stocks subject to mutual fund buying pressure in quarter t=0. In Panel A, abnormal stock returns are returns in excess of the returns of all stocks held by mutual funds that quarter. In Panel B, abnormal stock returns are industry-adjusted returns using the Fama-French 48 industry classifications. Inflow-driven Pressure is buying pressure by mutual funds experiencing large capital inflows, but not by other mutual funds. Non-Inflow-driven Pressure is widespread buying pressure by mutual funds other than those experiencing large capital flows. The Inflow-driven Pressure sample consists of 1733 stock-quarters in the top decile of Pressure in quarter t=0, but in the middle three deciles of UPressure in quarter t=0, between 1990 and 2002. The Non-Inflow-driven Pressure sample consists of 12,742 stock-quarters in the top decile of Upressure in quarter t=0. Pressure of stock i in quarter t is a stock-level measure of flow-motivated trading by all mutual funds j, and is calculated as Pressurei,t =

∑j

,0(max( ∆holdingj,i,t) | flowj,t>90th percentile) – ∑j

,0(max( −∆holdingj,i,t) | flowj,t<10th percentile)

Shares Outstandingi,t-1 Upressure is a measure of widespread trading by mutual funds that is not motivated by capital flows and is intended to capture information-motivated trading. The middle three deciles of Upressure capture stock quarters that are not subject to widespread net trading in any direction. Upressurei,t = {∑ ∆holding

jj,i,t | 10th percentile≤flowj,t≤90th percentile } / Shares Outstandingi,t-1

Mean abnormal returns are calculated each quarter for the portfolio of Inflow-driven Pressure stocks and Non-Inflow-driven Pressure stocks, and the time series of portfolio abnormal returns are used for statistical inference to control for cross-sectional correlation. *** (**) [*] represents one-tailed statistical significance at less than 1% (5%) [10%].

Panel A

Quarter Inflow-driven Pressure Non-Inflow-driven Pressure

t-4 5.29% *** 3.22% ***

t-3 6.29% *** 2.82% ***

t-2 5.15% *** 3.64% ***

t-1 6.90% *** 4.56% ***

t=0 7.70% *** 3.84% ***

t+1 -1.63% * 0.56% t+2 -1.17% -0.59% *

t+3 -2.58% ** -0.67% *

t+4 -3.15% ** -0.83% **

t+5 -1.97% ** -1.00% ***

t+6 -2.27% ** -0.11%

[t+1, t+6] -12.77% *** -2.65% *

28

Panel B: Industry-adjusted Returns

Quarter Inflow-driven Pressure Non-Inflow-driven Pressure

t-4 3.14% ** 2.82% ***

t-3 5.00% *** 2.64% ***

t-2 3.38% *** 3.31% ***

t-1 5.21% *** 3.84% ***

t=0 6.35% *** 3.04% ***

t+1 -2.91% ** 0.40% t+2 -1.62% -0.83% t+3 -2.56% ** -0.79% t+4 -1.99% * -0.59% t+5 -2.54% ** -0.90% *

t+6 -1.83% * -0.18%

[t+1, t+6] -13.46% *** -2.88% *

29

Table 3: Descriptive Statistics Panel A reports descriptive statistics for the full sample, Panel B reports descriptive statistics for the stocks subject to inflow-driven buying pressure (IBP sample) and Panel C reports descriptive statistics for stocks subject to widespread buying pressure by all mutual funds other than funds in the top decile of capital flows. ROA is operating income before depreciation over total assets at the end of the same quarter in the prior year. 1-year Price appreciation is defined as the growth in stock price over a one-year period ending at the end of the current quarter. BTM is book value of shareholders’ equity over market value of equity at the end of the same quarter in the prior year. Size is the natural logarithm of total assets at the end of the same quarter in the prior year. Leverage is long-term debt plus long-term debt in current liabilities, over total assets, at the end of the same quarter in the prior year. Dividend yield is the dividend per share divided by the stock price in the same quarter last year. Cash is cash and short-term investments over total assets at the end of the same quarter in the prior year. InsidSale (All) is the ratio of shares sold by all insiders to the sum of shares sold and purchased by all insiders in a firm-quarter. InsidSale (Execs) is the ratio of shares sold by the top 5 executives to the sum of shares sold and purchased by the top 5 executives in a firm-quarter. InsidHolding is the insider holdings, defined as shares held by insiders scaled by total shares outstanding. The sample consists of 239,112 stock-quarters from 1990 through 2002. Panel A: Full Sample Mean StdDev 1st Quartile Median 3rd QuartileROA 0.018 0.050 0.006 0.022 0.042 1-year Price appreciation 0.147 0.708 -0.265 0.029 0.367 BTM 0.699 0.555 0.322 0.567 0.910 Size 5.174 2.148 3.613 5.037 6.615 Leverage 0.210 0.198 0.042 0.162 0.324 Dividend yield 0.004 0.008 0.000 0.000 0.004 Cash 0.158 0.205 0.021 0.070 0.205 InsidSale (All) 0.382 0.373 0.000 0.321 0.685 InsidSale (Execs) 0.367 0.404 0.000 0.212 0.752 InsidHolding 0.027 0.070 0.000 0.003 0.019

Panel B: IBP Sample Mean StdDev 1st Quartile Median 3rd QuartileROA 0.032 0.050 0.013 0.032 0.054 1-year Price appreciation 0.530 1.042 -0.130 0.230 0.792 BTM 0.515 0.405 0.250 0.417 0.653 Size 5.897 1.642 4.742 5.658 6.845 Leverage 0.197 0.199 0.018 0.148 0.311 Dividend yield 0.003 0.007 0.000 0.000 0.003 Cash 0.207 0.243 0.023 0.101 0.305 InsidSale (All) 0.494 0.357 0.130 0.500 0.845 InsidSale (Execs) 0.471 0.392 0.000 0.493 0.912 InsidHolding 0.017 0.047 0.000 0.003 0.012

30

Panel C: Widespread Buying Pressure Sample Mean StdDev 1st Quartile Median 3rd QuartileROA 0.035 0.040 0.019 0.035 0.053 1-year Price appreciation 0.287 0.750 -0.152 0.130 0.515 BTM 0.527 0.394 0.265 0.439 0.682 Size 6.232 1.612 5.064 6.061 7.267 Leverage 0.217 0.199 0.033 0.185 0.339 Dividend yield 0.003 0.007 0.000 0.000 0.003 Cash 0.163 0.210 0.017 0.063 0.232 InsidSale (All) 0.458 0.354 0.070 0.480 0.758 InsidSale (Execs) 0.458 0.392 0.000 0.489 0.902 InsidHolding 0.016 0.047 0.000 0.002 0.009

31

Table 4: Timing Seasoned Equity Offerings The table reports coefficients and p-values from logit regressions of the equity issuance choice on the independent variables shown. The dependent variable is 1 if the firm has a seasoned equity offering (SEO) in quarter t, and zero otherwise. Buy Pressure (t−1, t−2) is a dummy that equals 1 if the stock was subject in quarter t−1 or quarter t−2 to severe buying pressure by mutual funds in the highest decile of capital flows but not subject to buying pressure by other mutual funds (inflow-driven buying pressure only). ROA is operating income before depreciation over total assets at the end of the same quarter in the prior year. 1-year Price appreciation is defined as the growth in stock price over a one-year period ending at the end of the current quarter. BTM is book value of shareholders’ equity over market value of equity at the end of the same quarter in the prior year. Size is the natural logarithm of total assets at the end of the same quarter in the prior year. Leverage is long-term debt plus long-term debt in current liabilities, over total assets, at the end of the same quarter in the prior year. Dividend yield is the dividend per share divided by the stock price in the same quarter last year. Cash is cash and short-term investments over total assets at the end of the same quarter in the prior year. The model includes year, quarter and industry fixed effects, and standard errors are clustered on both firm and time (two-dimensional clustering). The sample consists of 239,112 firm-quarters from 1990 through 2002. The odds ratio is the ratio of the odds of an SEO when Buy Pressure (t−1, t−2) =1 to the odds of an SEO when Buy Pressure (t−1, t−2) =0. *** (**) [*] represents one-tailed statistical significance at less than 1% (5%) [10%]. Indep. Variable Coefficient Odds Ratio Intercept -6.717 *** Buy Pressure (t-1, t-2) 0.374 *** 1.454 ROAt-4 -0.169 1-year Price appreciation 0.808 *** BTMt-4 -0.919 *** Sizet-4 0.195 *** Leveraget-4 0.622 *** Dividend yieldt-4 -24.305 *** Casht-4 0.753 *** Year f.e. Yes Quarter f.e. Yes Industry f.e. Yes R-square 10.59%

32

Table 5: Timing Insider Sales The table reports coefficients and p-values from regressions of insider sales (InsidSale) in quarter t on the independent variables shown. InsidSale is the ratio of shares sold by insiders in a firm-quarter to the sum of shares sold and purchased by insiders in that firm-quarter. The second and third columns report results for all insiders, while the last two columns report results for the top 5 executives. Buy Pressure (t−1, t−2) is a dummy that equals 1 if the stock was subject in quarter t−1 or quarter t−2 to severe buying pressure by mutual funds in the highest decile of capital flows but not subject to buying pressure by other mutual funds (inflow-driven buying pressure only). Size is the natural logarithm of total assets at the end of the same quarter in the prior year. 1-year Price appreciation is defined as the growth in stock price over a one-year period ending at the end of the current quarter. BTM is book value of shareholders’ equity over market value of equity at the end of the same quarter in the prior year. BTM is ranked into deciles from 1 (low BTM) to 10 (high BTM) and the decile dummies are denoted BTM1, BTM2 and so on. The BTM10 dummy is omitted. InsidHolding is the insider holdings, defined as shares held by insiders scaled by total shares outstanding. The model includes year, quarter and industry fixed effects and standard errors are clustered on both firm and time (two-dimensional clustering). The All Insiders sample consists of 155,624 firm-quarters, and the Top 5 Execs sample consists of 81,016 firm-quarters, from 1990 through 2002. *** (**) [*] represents one-tailed statistical significance at less than 1% (5%) [10%].   All Insiders     Top 5 Execs             Indep. Variable Coefficient   Coefficient      Intercept 0.2008 ***   0.0769 *** Buy pressure (t-1,t-2) 0.0372 ***   0.0357 *** Size t-4 0.0100 ***   0.0165 *** InsidSale t-4 0.1634 ***   0.1964 *** 1-year Price appreciation 0.0745 ***   0.0834 *** BTM1 0.1183 ***   0.1767 *** BTM2 0.1040 ***   0.1600 *** BTM3 0.0965 ***   0.1388 *** BTM4 0.0848 ***   0.1221 *** BTM5 0.0728 ***   0.1064 *** BTM6 0.0635 ***   0.0852 *** BTM7 0.0423 ***   0.0630 *** BTM8 0.0232 ***   0.0478 *** BTM9 0.0321 ***   0.0409 *** InsidHolding t 0.3420 ***   1.1337 *** Year f.e. Yes   Yes Quarter f.e. Yes   Yes Industry f.e. Yes   Yes R-square 8.34%   13.29%

33

Table 6: Future Performance of Timed SEOs This table reports the alpha (intercept) and coefficients from calendar-time Fama and French (1993) three-factor regressions. The dependent variable is the monthly excess return, over the risk-free rate, on a portfolio of all Timed SEOs, non-timed SEOs (Other SEOs) or stocks with no SEO but subject to buying pressure only by mutual funds in the highest decile of capital flows (Non-SEO Buy-Pressured). The stocks are held for 3 years (Panel A) or 5 years (Panel B). Mkt-Rf, SMB and HML are the Fama-French factors. Timed SEOs are SEOs within two quarters after the stock is subject to flow-driven buying pressure by mutual funds, while non-timed SEOs are all Other SEOs. Non-SEO Buy-Pressured stocks are those subject within the last two quarters to buying pressure by mutual funds with large capital inflows but not by other mutual funds (inflow-driven buying pressure only). The sample includes 3912 non-utility SEOs between 1990 and 2002, with 69 Timed SEOs. There are 1664 Non-SEO Buy-Pressured stock-quarters. The alpha is in percentage points. *** (**) [*] represents one-tailed significance at 1% (5%) [10%]. Panel A: 3 year Timed SEOs Other SEOs Non-SEO Buy-Pressured α -0.82 * -0.57 *** -0.34 *

Mkt-Rf 1.57 *** 1.32 *** 1.32 ***

SMB 0.50 ** 0.82 *** 0.61 ***

HML -0.01 -0.08 0.15 Adj. Rsq 62% 89% 79%

Panel B: 5 year       Timed SEOs Other SEOs Non-SEO Buy-Pressured   α -0.73 ** -0.29 ** -0.27 *

Mkt-Rf 1.55 *** 1.26 *** 1.27 ***

SMB 0.69 *** 0.86 *** 0.61 ***

HML 0.11 -0.01 0.16 Adj. Rsq 70% 90% 80%

34

Stock Price Pressure Due To Mutual Fund Buying Pressure

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

t-4 t-3 t-2 t-1 t=0 t+1 t+2 t+3 t+4 t+5 t+6

Event Quarter (t=0 is Mutual Fund Buying Quarter)

Cum

ulat

ive

Ave

rage

Abn

orm

al

Ret

urns

Inflow-driven BuyingPressureNon-Inflow-driven BuyingPressure

Figure 1: The figure shows cumulative average abnormal returns of stocks subject to severe buying pressure by mutual funds. Abnormal returns are returns in excess of the returns of all stocks held by mutual funds that quarter. We sum average quarterly abnormal returns to obtain the cumulative average abnormal returns. Inflow-driven Buying Pressure stocks are subject to buying pressure by mutual funds experiencing large capital inflows but not by other mutual funds. Non-Inflow-driven Buying Pressure stocks are subject to widespread buying pressure by mutual funds other than those experiencing large capital inflows. The Flow-driven Buying Pressure sample consists of 1733 stock-quarters in the top decile of Pressure in quarter t=0, but in the middle three deciles of UPressure in quarter t=0, between 1990 and 2002. The Non-Inflow-driven Buying Pressure sample consists of 12,742 stock-quarters in the top decile of UPressure in quarter t=0. Pressure of stock i in quarter t is a stock-level measure of flow-motivated trading by all mutual funds j, and is calculated as Pressurei,t =

∑j

,0(max( ∆holdingj,i,t) | flowj,t>90th percentile) – ∑j

,0(max( −∆holdingj,i,t) | flowj,t<10th percentile)

Shares Outstandingi,t-1 UPressure is a measure of widespread trading by mutual funds that is not motivated by capital flows and is intended to capture information-motivated trading. The middle three deciles of UPressure capture stock quarters that are not subject to widespread net trading in any direction. UPressurei,t = { ∆holding∑

jj,i,t | 10th percentile≤flowj,t≤90th percentile } / Shares Outstandingi,t-1

35

Stock Price Pressure Due To Mutual Fund Buying Pressure: Industry-Adjusted Returns

0

0.05

0.1

0.15

0.2

0.25

t-4 t-3 t-2 t-1 t=0 t+1 t+2 t+3 t+4 t+5 t+6

Event Quarter (t=0 is Mutual Fund Buying Quarter)

Cum

ulat

ive

Ave

rage

Abn

orm

al

Ret

urns

Inflow-driven BuyingPressureNon-Inflow-driven BuyingPressure

Figure 2: The figure shows cumulative average abnormal returns of stocks subject to severe buying pressure by mutual funds. Abnormal returns are industry-adjusted returns, using the Fama-French 48 industry classifications. We sum average quarterly abnormal returns to obtain the cumulative average abnormal returns. Inflow-driven Buying Pressure stocks are subject to buying pressure by mutual funds experiencing large capital inflows but not by other mutual funds. Non-Inflow-driven Buying Pressure stocks are subject to widespread buying pressure by mutual funds other than those experiencing large capital inflows. The Flow-driven Buying Pressure sample consists of 1733 stock-quarters in the top decile of Pressure in quarter t=0, but in the middle three deciles of UPressure in quarter t=0, between 1990 and 2002. The Non-Inflow-driven Buying Pressure sample consists of 12,742 stock-quarters in the top decile of UPressure in quarter t=0. Pressure of stock i in quarter t is a stock-level measure of flow-motivated trading by all mutual funds j, and is calculated as Pressurei,t =

∑j

,0(max( ∆holdingj,i,t) | flowj,t>90th percentile) – ∑j

,0(max( −∆holdingj,i,t) | flowj,t<10th percentile)

Shares Outstandingi,t-1 UPressure is a measure of widespread trading by mutual funds that is not motivated by capital flows and is intended to capture information-motivated trading. The middle three deciles of UPressure capture stock quarters that are not subject to widespread net trading in any direction. UPressurei,t = { ∆holding∑

jj,i,t | 10th percentile≤flowj,t≤90th percentile } / Shares Outstandingi,t-1

36