The Real Consequences of Market Segmentation∗

Sergey Chernenko Adi Sunderam

The Ohio State University Harvard University

Fisher College of Business

chernenko.1@osu.edu sunderam@fas.harvard.edu

August 3, 2010

Abstract

We study the real effects of market segmentation due to credit ratings using a

matched sample of firms just above and just below the investment-grade cutoff. These

firms have similar observables, including average investment rates. However, flows into

high-yield mutual funds have an economically significant effect on the issuance and

investment of the speculative-grade firms relative to their matches, especially for firms

likely to be financially constrained. The effect is associated with the discrete change
∗
We thank Malcolm Baker, Bo Becker, Effi Benmelech, Dan Bergstresser, Alexander Butler, John Camp-
bell, Lauren Cohen, Andrew Ellul, Fritz Foley, Robin Greenwood, Sam Hanson, Victoria Ivashina, Andrew
Karolyi, Darren Kisgen, David Scharfstein, Erik Stafford, Jeremy Stein, and seminar participants at Arizona
State University, the Bank of Canada, Cornell, the Federal Reserve Bank of Boston, the Federal Reserve Bank
of New York, Harvard, the Finance Down Under Conference, Imperial College London, the London School
of Economics, the NBER Credit Rating Agencies session, Ohio State University, Rice, the Trans-Atlantic
Doctoral Conference, the University of Illinois at Urbana-Champaign, the WFA 2010 Annual Meetings,
and Yale for helpful comments and suggestions, Doug Richardson at the Investment Company Institute for
providing mutual fund flows data, and Jerome Fons, Martin Fridson, Mark Mitchell, Oleg Melentyev, and
Michael Weilheimer for discussing the high-yield market with us.
in label from investment-grade to speculative-grade, not with changes in continuous

measures of credit quality. We do not find similar effects at other rating boundaries.

JEL Classifications: G24, G31

Key Words: credit ratings, firm investment, mutual fund flows, market segmentation,

comovement
 Capital markets play a critical role in the efficient allocation of capital across firms.
Their ability to play this role, however, may be impeded by market segmentation. In this
paper, we study one of the most prominent divides in capital markets—the distinction drawn
between investment- and speculative-grade firms. A large number of regulations, investment
charters, and contracts reference this distinction, and recent research suggests that it can
affect firms’ capital structure and cost of capital (Kisgen, 2006; Kisgen and Strahan, 2009;
Ellul, Jotikasthira, and Lundblad, 2010).
Market segmentation between investment- and speculative-grade firms may also have real
effects. When retail investors withdraw capital from high-yield mutual funds, which are large
buyers of speculative-grade bonds, arbitrage capital may not immediately offset this shock
(Mitchell, Pedersen, and Pulvino, 2007; Duffie and Strulovici, 2009). As a result, it may
affect the supply and cost of capital available to speculative-grade firms. This in turn may
cause some firms, particularly those unable to access other sources of financing, to cut their
investment. This paper presents evidence of this mechanism at work.
Simple comparisons between investment- and speculative-grade firms would be con-
founded by differences in fundamentals between them. Instead, drawing on the econometric
literature on treatment effects and regression discontinuity, we construct a matched sample
of firms just above (BBB-) and just below (BB+) the investment-grade cutoff. These firms
are similar on observable characteristics and have the same average rates of investment, but
high-yield fund flows only affect the supply and cost of capital for the speculative-grade
(BB+) firms. As a result, fund flows cause the investment of firms just below the cutoff to
diverge from the investment of their matches just above the cutoff. This effect is economi-
cally meaningful—a one standard deviation increase in fund flows increases the investment
of BB+ firms relative to their BBB- matches by about 7% of the average rate of investment.

1
The effect is stronger for smaller, financially constrained firms that have limited ability to
substitute to either bank loans or the asset-backed securities market, as well as firms that
depend on external financing. Our results indicate that market segmentation causes tempo-
rary differences in the investment of similar firms, but that these differences tend to average
out over time.
In principle, these results could be driven by an increase in the availability of financing
or a decrease in the cost of capital. We explore the mechanism driving our results, showing
that fund flows increase the bond issuance of BB+ firms relative to their BBB- matches.
Again, the effect tends to be stronger for financially constrained firms. Thus, while we cannot
rule out pure cost of capital effects, our results suggest that the availability of capital is an
important driver of investment.
A natural objection to our interpretation of these results is that fund flows may be antici-
pating investment opportunities. We match BB+ and BBB- firms based on industry and firm
characteristics such as size, leverage, Altman’s z-score, and Q. Our identifying assumption
is that firms close to the cutoff with similar observable characteristics are subject to similar
shocks to profitability and investment opportunities. If this is the case, by differencing the
investment rates of matched BB+ and BBB- firms we difference out any common shocks
that may be correlated with fund flows. We can then interpret the differential effect of fund
flows on the investment of BB+ firms as evidence of a supply of capital effect.
Despite our matching methodology, the investment opportunities of matched firms may
still not be exactly the same, and fund flows may be picking up the differential investment
opportunities of less creditworthy firms. We address this concern in a number of ways. First,
drawing on the regression discontinuity literature, we show that our results are robust to
controlling for interactions of fund flows with continuous measures of credit quality, including

2
size, leverage, and Altman’s z-score. Second, we show that our results are robust to con-
trolling for a variety of macroeconomic variables, including the VIX, the term spread, the
Baa-Aaa credit spread, and aggregate stock market returns. Third, we conduct falsification
tests and find that the investment-grade cutoff is the only one where the investment of firms
below the cutoff is more sensitive to fund flows than the investment of firms above the cutoff.
Moreover, BB+ firms constitute only 10% of all speculative-grade firms, so fund flows are
likely to be driven by the performance and investment prospects of lower-rated firms, which
are very different from BB+ firms on observable characteristics. This suggests that our re-
sults may be due to categorization. If investors emphasize differences between investment-
and speculative-grade firms and overlook differences within these broad categories, they may
treat two firms in the same category similarly, even if the firms have very different perfor-
mance and investment prospects (Mullainathan, 2000; Barberis and Shleifer, 2003). Our
evidence is consistent with this idea. We find that even after controlling for the invest-
ment of matched BBB- firms, the investment of BB+ firms comoves with the investment of
lower-rated speculative-grade firms in unrelated industries.
Another objection to our identification strategy is that firms could be selecting into
different ratings based on unobservable characteristics.1 The direction of any bias introduced
by selection on unobservable characteristics is ambiguous. While it could be the case that
firms with higher sensitivities of investment to high-yield fund flows choose to be speculative
grade, we believe that the more natural selection story goes the opposite way. Firms whose
investment would be most affected by the volatility of fund flows if they were rated BB+
have the strongest incentives to alter their behavior to achieve a BBB- rating. Thus, firms

1
The distribution of credit ratings is reported in Appendix Figure 2 and suggests that some selection might
be taking place, as there are fewer firm-quarter observations rated BB+ than either BBB- or BB.

3
that do carry a BB+ rating in our data are likely to have a relatively low sensitivity of
investment to fund flows.
In summary, we find that shocks to the supply of capital of high-yield mutual funds cause
the investment of firms just below the investment-grade cutoff to diverge from the investment
of similar firms just above the cutoff. Our work contributes to the literature studying the
role of credit ratings in capital markets.2 Most papers in this literature study the information
content of credit ratings and their effects on assets prices and capital structure, while we focus
on real investment. Lemmon and Roberts (2009) are perhaps the most closely related to
our work. They investigate the effect on firm investment of the collapse of Drexel Burnham
Lambert and of the concurrent regulatory changes. They use a differences-in-differences
approach to compare the investment of all speculative-grade firms with unrated firms. We
compare the financing and investment behavior of firms on both sides of the investment-grade
cutoff, and suggest that the effect of fund flows on the investment of BB+ firms just below
the cutoff is a spillover effect due to categorization. Furthermore, while the existing literature
focuses on isolated changes in the institutional environment, we study how recurring shocks
to the capital of an important investor class interact with market segmentation to affect real
investment.
2
Harold (1938) and Hickman (1958) are among the first to study the use of credit ratings and in particular
their ability to forecast defaults. The papers collected in Levich, Majnoni, and Reinhart (2002) study the
various aspects of the role of credit ratings in the global financial markets. More recently, Faulkender and
Petersen (2006) find that firms with a credit rating that allows them to access public debt markets have 35%
more debt than other similar firms. Kisgen (2006) argues that firm financing decisions are affected by the
discrete costs and benefits of different credit ratings. Sufi (2009) studies how the introduction of syndicated
bank loan ratings by Moody’s and Standard & Poor’s in 1995 expanded the set of investors able to invest in
syndicated loans and led to increased debt issuance and investment by lower-rated borrowers. Kisgen and
Strahan (2009) study the recent certification of ratings from Dominion Bond Rating Service for regulatory
purposes. They find that after certification bond yields fell for firms for whom Dominion had a higher rating
than the other certified rating agencies. Ellul, Jotikasthira, and Lundblad (2010) argue that regulatory
constraints can force insurance companies to sell recently downgraded bonds, generating downward price
pressure and subsequent reversals.

4
 In studying the capital supply effects of high-yield fund flows, our work also contributes to
the growing literature on market-driven corporate finance, reviewed recently by Baker (2009).
With the exception of the papers studying the bank lending channel,3 most of the literature
relating firm investment to the supply of capital focuses on the equity markets.4 Since for
most firms external equity is a relatively minor source of financing compared to retained
earnings and debt (Mayer, 1988), it is important to understand the effects of market-driven
corporate finance in credit markets.
The remainder of the paper is organized as follows. In the next section we review the
institutional background that motivates our empirical methodology, which we discuss in
more detail in Section II. Section III describes our data and summarizes differences in firm
characteristics across credit ratings. Section IV reports our main results, and Section V
concludes.

I Institutional background

We begin by briefly describing two institutional features of credit markets and credit
ratings that motivate our empirical methodology. First, many regulatory and voluntary con-
ventions restrict the ability of different investor groups to hold speculative-grade securities.
Second, rating methodologies introduce noise and inertia in credit ratings. Once we review
this institutional background, we go on to describe our empirical methodology.

3
See, for example, Bernanke and Blinder (1992), Kashyap, Stein, and Wilcox (1993), Gertler and Gilchrist
(1994), Bernanke and Gertler (1995), Kashyap and Stein (2000), and Leary (2009).
4
See, for example, Baker, Stein, and Wurgler (2003) and Polk and Sapienza (2009).

5
A Regulations restrict holdings of speculative-grade securities

Many rules and regulations restrict the ability of different investor groups to hold speculative-
grade securities. Commercial banks have been prohibited from holding bonds rated BB+
and below since 1936.5 The Financial Institutions Reform, Recovery, and Enforcement Act
of 1989 extended the ban on commercial bank holdings of speculative-grade bonds to thrifts.6
Most state insurance regulations follow the guidelines established by the National Associ-
ation of Insurance Commissioners, which sets both higher risk charges for and a hard cap
on holdings of speculative-grade bonds.7 In addition, the net capital rule for broker-dealers
requires larger haircuts for speculative-grade securities (U.S. Securities and Exchange Com-
mission, 2003).8
Although not subject to regulatory restrictions, most bond mutual funds specialize in
either investment- or speculative-grade bonds. Investment-grade funds typically limit their
holdings of speculative-grade bonds to 5-10% of fund assets. For instance, PIMCO Total
Return Fund, the largest corporate bond fund, limits its holdings of high-yield bonds to 10%
of fund assets (PIMCO Total Return Fund Prospectus). As of June 2009, PIMCO Total
Return Fund invested 5% of its assets in speculative-grade and unrated securities. Vanguard

5
West (1973) argues that these regulations increased the spreads for speculative-grade borrowers relative to
investment-grade ones.
6
The act prohibited purchases of speculative-grade bonds and mandated existing holdings to be liquidated
by 1994. As a result, thrifts’ share of the corporate bond market fell from around 7% in 1988 to 1% by 2008
(Flow of Funds Accounts of the United States, Table L212, Corporate and Foreign Bonds).
7
Risk charges for A, BBB, BB, and B rated bonds are 0.4%, 1.3%, 4.6%, and 10% respectively. The
portfolio share of all non-investment grade bonds is capped at 20%. As a result of these restrictions,
insurance companies’ share of all speculative-grade bonds is only 8.5%, one-fourth of their 34% share of all
investment-grade bonds (Ellul, Jotikasthira, and Lundblad, 2010).
8
Haircuts for investment-grade nonconvertible debt securities paying a fixed interest rate vary between 2%
and 9% depending on maturity. Haircuts for speculative-grade bonds are generally 15%.

6
Short-, Intermediate-, and Long-Term Investment-Grade Funds are more restrictive and in-
vest exclusively in securities rated BBB- and above (Vanguard Bond Funds Prospectus).
Some investment-grade funds, such as Fidelity U.S. Bond Index Fund, do not have explicit
limits on the credit quality of their portfolio holdings. Instead, they seek “to provide invest-
ment results that correspond to the total return of the bonds in the Lehman Brothers U.S.
Aggregate Index” (Fidelity U.S. Bond Index Fund Prospectus). Since the index consists
exclusively of investment-grade securities, any speculative-grade holdings would expose such
funds to significant tracking error and would thus be quite costly. Not surprisingly, as of June
2009, Fidelity U.S. Bond Index Fund invested only 0.5% of its assets in speculative-grade
bonds.
Conversely, high-yield funds specify a minimum share of their assets to be invested in
speculative-grade securities. Vanguard High-Yield Corporate Fund must invest “at least
80% of its assets in corporate bonds that are rated below Baa by Moody’s Investor Service,
Inc. (Moody’s); have an equivalent rating by any other independent bond-rating agency”
(Vanguard High-Yield Corporate Fund Prospectus). As of June 2009, Vanguard High-Yield
Corporate Fund held 91% of its bond portfolio in speculative-grade and unrated bonds. The
80% minimum on holdings of speculative-grade bonds is typical—as of June 2009, high-yield
mutual funds on average had 87% of their bond portfolios invested in speculative-grade
bonds.
Finally, note that speculative-grade bond purchases by investment-grade funds and investment-
grade bond purchases by high-yield funds will introduce noise in our measure of the capital
supply available to speculative-grade firms, and bias us against finding any results.

7
B The muddled origins of “investment grade”

Given the large number of restrictions on investing in speculative-grade securities, one

may worry that the specific location of the investment-grade cutoff between BBB- and BB+
is endogenous. In particular, differences in firm characteristics and investment opportunities
may be especially stark across this ratings transition. We examine differences in observ-
able firm characteristics below, but the origins of the distinction between investment and
speculative grades may also mitigate some of these concerns.9
When Moody’s published the first credit ratings in 1909, it did not use the terms “invest-
ment grade” and “speculative grade.” Instead the agency used the term “grade” to refer to
three groups of credit ratings: AAA, AA, and A bonds constituted the “first-grade,” BBB
and BB bonds the “second-grade,” B and lower rated bonds “low grade.” Thus in contrast
to the modern distinction between BBB and BB bonds, Moody’s originally thought of them
as being of similar quality.10
It was not until 1931 that the modern distinction between speculative- and investment-
grade bonds began to emerge. On September 11, 1931, the Comptroller of the Currency ruled
that commercial banks could carry bonds rated BBB or higher at cost, but that they had to
mark to market lower rated and defaulted bonds.11 On February 15, 1936, the Comptroller
and the Federal Reserve went further and completely prohibited commercial banks from
purchasing “ ‘investment securities’ in which the investment characteristics are distinctly or
9
See Harold (1938) and Fons (2004) for more on the history of credit ratings and the distinction between
investment and speculative grades.
10
As Poor’s Publishing, Standard Statistics, and Fitch Publishing entered the credit ratings market in
1916, 1922, and 1924, they generally followed similar characterizations. All agencies described BB as either
“Good” or “Fair,” and none referred to it as speculative (Harold, 1938).
11
Although the ruling applied only to national banks, many state banking regulators followed the
Comptroller’s lead and introduced similar restrictions for state chartered banks.

8
predominantly speculative” (Harold, 1938).
The ruling caused significant confusion regarding the precise definition of “speculative”
securities. American Banker concluded that the “regulation limits investments practically
to those with an A rating” (Harold, 1938). Faced with an outcry from the banking com-
munity, the Comptroller backtracked, and by 1938 Moody’s persuaded the regulators that
bonds rated BBB were not “distinctly or predominantly speculative.” Subsequent regula-
tions continued to use the same BBB vs. BB cutoff, but its origins suggest that the cutoff
could have just as easily been drawn at A vs. BBB or BB vs. B.
This history suggests that the investment-grade cutoff was not originally drawn to dis-
tinguish between firms with sharply different fundamentals. Over time, however, market
institutions may have evolved around the cutoff to render its location more correlated with
firm characteristics and investment opportunities. Below we provide evidence that this is not
the case, showing that differences in observable firm characteristics at the investment-grade
cutoff are similar to differences across other rating cutoffs.

C Noise and inertia in credit ratings

Credit ratings do carry information about firms’ credit quality and potentially about
their investment opportunities.12 However, if ratings are subject to noise and inertia, we will
be able to find pairs of firms that are similar on observable characteristics but are on different
sides of the cutoff. There are a number of reasons to believe that credit ratings are noisy,
12
There is still considerable debate, however, as to whether credit ratings contain any information not already
available to investors. Market-based measures of credit quality tend to be better predictors of default than
credit ratings, at least at short- and medium-term horizons (Cantor and Mann, 2006). Kliger and Sarig (2000)
and Tang (2009) argue that Moody’s refinement of its rating system revealed new information about rated
firms. Jorion, Liu, and Shi (2005) find greater informational effects of credit rating changes after Regulation
Fair Disclosure prohibited companies from selectively disclosing nonpublic information, but excluded rating
analysts from the new regulation.

9
lagging measures of credit quality. First, ratings methodologies emphasize stability. The
agencies explicitly trade off rating accuracy versus stability and are reluctant to upgrade or
downgrade firms if such changes might have to be reversed in the future (Cantor and Mann,
2006). This is particularly true at the investment-grade cutoff, as the agencies are aware
that their decisions affect the ability of market participants to hold certain bonds. Moreover,
even when the agencies do adjust credit ratings, the adjustment is likely to be only partial,
followed by additional changes (Altman and Kao, 1992). As a result, market-based measures
such as yield spreads are more accurate than credit ratings in forecasting defaults at short-
and medium-term horizons (Cantor and Mann, 2006).13
In addition to the inertia in ratings generated by the explicit goal of stability, credit rating
agencies’ organizational structures may create incentives for analysts to be conservative
in upgrading or downgrading firms. One such organizational practice, used by Moody’s
Leveraged Finance Group, is having separate groups analyzing investment- and speculative-
grade credits. This organizational structure could create conflicts of interest as the group
covering a particular firm would lose fee revenue if it upgraded or downgraded the client
across the investment-grade cutoff.
Rating outlooks introduce a further wedge between the information and regulatory con-
tent of credit ratings. These outlooks “assess the potential for an issuer rating change” but
are not “necessarily a precursor for a rating change” (Standard & Poor’s, 2008). On average,
issuers with a positive outlook default at the same rate as issuers rated one notch higher
13
Appendix Table XVI provides some suggestive evidence of credit ratings inertia in our data. Panel A
compares the characteristics of BB+ firms in the two years before they are upgraded with the characteristics
of all BBB- firms. Other than being smaller, BB+ firms that are subsequently upgraded perform much better
than an average BBB- firm. They have higher valuations, lower market leverage, and higher profitability
and cash holdings. Conversely, Panel B shows that BBB- firms that are downgraded in the next two years
perform significantly worse than the average BB+ firm. They have significantly higher leverage, lower interest
coverage, lower profitability, and they invest less than BB+ firms.

10
(Cantor and Hamilton, 2005).14 Thus, although the information content of a BB+ rating
with a positive outlook is the same as, if not better than, that of an unconditional BBB-
rating, their regulatory implications are quite different—until they are actually upgraded,
BB+ firms with positive outlooks cannot access the investment-grade market.

II Empirical methodology

The previous section suggests two stylized facts that drive our empirical approach. First,
regulations and investment charters restrict the ability of many investor groups to hold
speculative-grade securities. Thus, if there are limits to arbitrage, arbitrageurs might be un-
able to fully offset liquidity or noise shocks suffered by traditional speculative-grade investor
groups. In particular, since high-yield mutual funds hold about 20% of speculative-grade
bonds, flows into high-yield mutual funds can have a significant effect on the investment of
speculative-grade firms. Second, noise and inertia in credit ratings imply that there are BB+
firms that are similar to BBB- in terms of observable firm characteristics and investment
opportunities. That is, there are firms rated BB+ that “should be” BBB- and vice versa.
Our empirical strategy is to match BB+ and BBB- firms based on industry and firm
characteristics such as size, leverage, Altman’s z-score, and Q, and to compare the invest-
ment sensitivities of the matched firms to high-yield fund flows. Our benchmark matching
procedure uses industry and size.15 Within each industry-quarter we match each firm rated

14
In fact, over the 1995-2005 period, BB+ firms with a positive outlook had a 5-year default rate of
only 0.95%, significantly lower than the 3.88% default rate of BBB- firms with stable outlook (Cantor and
Hamilton, 2005).
15
Below we show that similar results are obtained when matching on leverage, z-score, or Q in addition to
size. We use size as our primary matching variable because it has the most power in explaining investment-
versus speculative-grade status, and because empirically it is an important determinant of investment.

11
BB+ with a firm rated BBB- that is of similar size. We consider firms to be of similar size
if the ratio of their book assets is within [0.5, 2]. We start looking for a match of similar
size within the Fama-French 48 industries, and if we do not find one we successively use the
coarser 38, 30, and 17 industry definitions until we either have a successful match or are
unable to match a given BB+ firm.
In addition to our benchmark matched sample, we examine two subsets of firms closer to
the investment-grade cutoff. The first subset consists of less levered BB+ firms. Excluding
the top 25% of BB+ firms by market leverage results in a matched sample of BB+ and BBB-
firms with virtually no differences in observable characteristics. The second subset consists
of BB+ firms with positive outlooks. These firms have observable characteristics and default
rates similar to BBB- firms, but they are still subject to shocks associated with high-yield
fund flows.
Our approach is similar in spirit to the pseudo-experimental approaches like regression
discontinuity and differences-in-differences. Consider a firm i with true (continuous measure
of) credit quality Si and assigned credit rating Ri . Its investment can be written as a function
of standard investment regression controls and flows into high-yield mutual funds:

Invi,t = αi +βQ ·Qi,t−1 +βCF ·CFi,t +βF lows (Ri )·F lowst−1 +InvOpportunitiest (Si )+εi,t . (1)

We assume that unobservable common shocks to investment opportunities InvOpportunitiest (Si )

vary continuously with the true credit quality Si , while the sensitivity of investment to fund
flows βF lows (Ri ) depends on the assigned rating because of the regulatory frictions described
above.16 Individual firms are too small for aggregate fund flows to be correlated with the
16
We are also assuming the coefficients on Q and cash flow to be the same across ratings. In untabulated
results we show that the assumption that βQ (BB+) = βQ (BBB−) and βCF (BB+) = βCF (BBB−) cannot
be rejected in our data. Our results are robust to allowing the coefficients on Q and cash flow, as well as

12
firm-specific shocks εi,t . Fund flows are potentially correlated with the common shocks
InvOpportunitiest , which would bias the coefficient βF lows upward if we ran the simple re-
gression above. However, if our matching procedure is effective, we can find a BBB- firm j
that has very similar underlying credit quality, Sj , and hence is subject to the same common
investment opportunities shocks, i.e. InvOpportunitiest (Si ) = InvOpportunitiest (Sj ). We
can then difference the two equations to obtain

Invi,t − Invj,t = (αi − αj ) + (βF lows (BB+) − βF lows (BBB−)) · F lowst−1

+ βQ · (Qi,t−1 − Qj,t−1 ) + βCF · (CFi,t − CFj,t ) + (εi,t − εj,t )

or more compactly

∆Invi,t = α + βF lows · F lowst−1 + βQ · ∆Qi,t−1 + βCF · ∆CFi,t + ηi,t

(2)

where ∆X = X BB+ − X BBB− is the difference in firm characteristic X between matched

BB+ and BBB- firms. By differencing the investment of matched firms, we thus remove
any correlation between fund flows and investment opportunities. Finding a positive and
statistically significant coefficient βF lows is then evidence of a capital supply effect of fund
flows on the investment of BB+ firms.17
other controls, to be different.
17
Note that the reasons we expect high-yield fund flows to affect the investment of speculative-grade firms do
not extend to investment-grade fund flows. Investment-grade funds account for a much smaller share of the
investment-grade corporate bond market than the share of high-yield mutual funds in the speculative-grade
bond market. Furthermore, investment-grade firms have access to more alternative sources of financing.
Therefore, we do not expect investment-grade flows to have any effect on the difference in the investment
rates of BB+ and BBB- firms, and including investment-grade flows would only make the statistical inference
more difficult. Untabulated results indicate that when investment- and speculative-grade fund flows are
included together, the coefficient on investment-grade fund flows is small and not statistically significant.

13
 Our matching procedure selects firms that have similar observable characteristics but are
on different sides of the investment-grade cutoff, and documents the differential sensitivity
of the investment of BB+ firms to fund flows. An alternative econometric approach to
implement our empirical strategy is propensity score weighting, which uses panel regression
employing the full sample of all BB+ and BBB- firms but gives more weight to firms close to
the cutoff (Hirano, Imbens, and Ridder, 2003; Imbens and Wooldridge, 2009). This approach
delivers similar results, which are reported in Appendix Table XIX.
Both matching and propensity score weighting assume that BB+ and BBB- firms are
subject to the same investment opportunities shocks. One might still be concerned that
investment opportunities are not quite the same and that flows are picking up the differential
investment opportunities of less creditworthy firms. We address this concern in a number
of ways. First, drawing on the regression discontinuity design literature,18 we include the
interactions of firm characteristics with fund flows to show that it is the discontinuous change
in credit rating and not changes in continuous measures of credit quality that determines
the sensitivity of firm investment to fund flows. Controlling for up to five powers of firm
characteristics interacted with flows, there is still a discontinuous change in the sensitivity
of investment to fund flows as firms are rated BBB- versus BB+.
Second, we show that our results are robust to controlling for a variety of macroeconomic
variables, including the VIX, the term spread, a credit spread, and aggregate stock market
returns. This helps to rule out the story that high yield fund flows are simply proxying for
the state of the macroeconomy, and that the investment opportunities of less creditworthy

The coefficient on speculative-grade fund flows retains its magnitude; its statistical signficance is somewhat
lower due to the correlation between investment- and speculative-grade fund flows.
18
See for example Imbens and Lemieux (2008).

14
firms are procyclical.
Third, the basic version of the differential investment opportunities story predicts that the
investment of lower-rated firms around any rating cutoff is more sensitive to fund flows than
the investment of higher-rated firms. For example, it predicts that the investment of B+ firms
is more sensitive to fund flows than the investment of BB- firms. This is not what we find in
falsification tests comparing the sensitivity of investment to fund flows around other rating
cutoffs—the investment-grade cutoff is the only one where the investment of firms below the
cutoff is more sensitive to fund flows than the investment of firms above the cutoff. Therefore,
one requires a more complicated story where the nature of investment opportunities changes
dramatically at only the investment grade cutoff. Both the characteristics of matched firms
and the historical origins of the investment-grade cutoff suggest that this is not the case.

III Data

In this section we describe our sample construction and variable definitions and address
three data-related issues before turning to our results. First, we discuss which of a firm’s
potentially numerous credit ratings determines whether it can access the investment-grade
market. Second, we examine differences in firm characteristics across credit ratings to show
that there is no abrupt change in firm characteristics around the investment-grade cutoff.
Third, we explain how we measure fund flows, and show that flows are large relative to the
capital and investment of speculative-grade firms.

15
A Sample construction

Our sample is the domestic firms in the quarterly CRSP/Compustat merged data set,
excluding financials and utilities and covering the 1986Q1-2007Q4 period. The sample pe-
riod is determined by the availability of S&P domestic long-term issuer credit ratings in
Compustat starting in December 1985. Some of our specifications use rating outlooks and
bank loan ratings from the S&P RatingsDirect and Ratings IQuery databases.
CAP Xi,t
We measure investment as P P Ei,t−1
, the ratio of capital expenditures in quarter t to net
property, plant, and equipment at the end of quarter t−1.19 Our regressions include standard
CFi,t
controls: cash flow normalized by lagged capital, P P Ei,t−1
, and Qi,t−1 . Cash flow is income
before extraordinary items plus depreciation.20 Q is the market value of equity from CRSP
plus assets minus the book value of stockholder equity, all divided by assets. We also control
for size, leverage, Altman’s z-score, and interest coverage. Market leverage is book debt
divided by the sum of book debt and market value of equity from CRSP. Altman’s z-score
is 1.2 · W Ci,t /Assetsi,t + 1.4 · REi,t /Assetsi,t + 3.3 · EBITi,t /Assetsi,t + Salesi,t /Assetsi,t ,
where W Ci,t is working capital, i.e., current assets minus current liabilities, and REi,t is
retained earnings (Altman, 1968). We use the modified version of Altman’s z-score, which
excludes leverage, because we control for leverage directly. Interest coverage is the ratio
of EBIT to interest expense, calculated using four-quarter moving averages of EBIT and

19
Because fund flows and investment could be linked through debt issuance, changes in assets could offset
the effect of flows on investment if we scaled capital expenditures by assets instead of PPE. Nevertheless, we
obtain similar but weaker results when we do so. Appendix Table XVII reports the results of our investment
regressions when investment is alternatively scaled by PPE, assets, or assets net of cash (to mitigate the
mechanical effect of debt issuance on the measured rate of investment). The results confirm our intuition—the
effect of fund flows on investment is strongest when investment is scaled by PPE, weakest when investment
is scaled by assets, and in between when investment is scaled by assets net of cash.
20
To make our results more comparable with papers using annual data, we annualize investment and cash
flow.

16
interest expense. To reduce the effect of outliers we winsorize all variables at the first and
ninety-ninth percentiles.

B Measuring access to the investment-grade market

While regulations distinguish between investment and speculative grades at the security
level, investment activity occurs at the firm level. We therefore need a firm-level measure
of access to the investment-grade market. The senior secured credit rating is typically the
highest rating a firm can achieve on an individual security and is therefore the right measure
of access to the investment-grade market. A firm with a BB+ senior secured rating has
no way to access the investment-grade market during periods of low or negative flows into
high-yield mutual funds.21 In comparison, a firm with a BB+ senior unsecured rating that
has unencumbered collateral may still be able to access investment-grade market by issuing
senior secured debt.
We use the S&P long-term issuer credit rating, which is a “current opinion of an issuer’s
overall creditworthiness, apart from its ability to repay individual obligations” and corre-
sponds closely to the senior secured rating (Standard & Poor’s, 2008). S&P may “notch
up”—rate individual issues above the issuer credit rating—when it “can confidently project
recovery prospects exceeding 70%” (Standard & Poor’s, 2008). Since few firms are in posi-
tion to issue senior secured bonds with recovery prospects exceeding 70%, the S&P long-term
issuer credit rating is a good measure of the firm’s ability to access the investment-grade
market.22 And to the extent that some firms with BB+ senior secured ratings are able to
21
The two primary exceptions are obtaining a guarantee from another entity and issuing asset-backed
securities.
22
During the 1982-2008 period, average recovery rates for senior secured bonds were 52.3% when measured by
post-default trading prices and 63.6% when measured by ultimate recoveries (Moody’s, 2009). In untabulated
results we estimate that less than 15% of all nonconvertible bond issues by non-financial firms are notched

17
issue higher rated securities, we will be less likely to find any effect of fund flows on the
investment of speculative-grade firms.

C No break in firm characteristics at the investment-grade cutoff

Our identification strategy and falsification tests require that differences in firm charac-
teristics at the investment-grade cutoff be similar to differences across other rating cutoffs.
Table I reports the means of firm characteristics by credit rating.23 As there are few AAA
and AA+ firms, we combine these firms into one category. We do the same for firms rated
CCC+ through CCC-.
Lower-rated firms are generally smaller and more levered. In addition, they have lower
values of Altman’s z-score and interest coverage than higher-rated firms. The ratio of net
property, plant, and equipment to assets is relatively constant across credit ratings. Q varies
from 2.5 for the most highly rated firms to 1.4 for CCC rated firms. Higher-rated firms are
more profitable than lower-rated firms, whether one looks at operating margins, ROA, or
cash flow. Despite significant differences in Q across credit ratings, firms appear to engage
in similar levels of capital expenditures.
Importantly, the investment-grade cutoff does not stand out compared to other rating
cutoffs. BB+ firms are on average 27.8% smaller than BBB- firms, but there are similar
differences in size around other lower rated cutoffs, and our empirical methodology matches
on size to produce a sample of comparably sized firms. The market leverage of BB+ firms is
11.6% larger than the market leverage of BBB- firms, but there are only three other cutoffs

up. When weighted by total proceeds, a bit more than 10% are notched up. The vast majority of notched
up issues are asset- and mortgage-backed bonds.
23
Appendix Table XV reports differences in firm characteristics across adjacent credit ratings.

18
 Table I
Means of Firm Characteristics by Credit Rating

This table reports the means of firm characteristics by credit rating for our sample of firms in the quarterly CRSP/Compustat
merged data set, excluding financials and utilities. The sample period is 1986Q1-2007Q4. Variable definitions are in the Appendix
Table XIII.
Investment Grade Speculative Grade
AA+ AA AA- A+ A A- BBB+ BBB BBB- BB+ BB BB- B+ B B- CCC

19
Assets 25306 18007 12077 11505 9908 8330 7639 5993 5253 3793 2362 1743 1243 1860 1421 1205
Book leverage 0.269 0.320 0.338 0.370 0.383 0.385 0.413 0.440 0.459 0.498 0.544 0.586 0.651 0.741 0.735 0.875
Market leverage 0.118 0.135 0.151 0.174 0.201 0.234 0.248 0.302 0.313 0.350 0.387 0.433 0.491 0.551 0.546 0.664
Z-score 1.293 1.193 1.183 1.176 1.083 0.983 0.890 0.851 0.813 0.734 0.687 0.578 0.433 0.071 −0.167 −0.462
Interest coverage 24.446 16.038 14.537 12.926 10.862 8.692 8.371 6.198 6.723 5.373 4.641 3.892 2.124 0.843 −0.664 −0.796
Cash/Assets 0.087 0.077 0.065 0.060 0.067 0.059 0.058 0.053 0.063 0.065 0.069 0.080 0.094 0.125 0.163 0.106
PPE/Assets 0.391 0.409 0.365 0.366 0.356 0.391 0.409 0.388 0.350 0.381 0.360 0.353 0.344 0.350 0.342 0.403
Q 2.477 2.411 2.201 2.076 1.939 1.778 1.751 1.520 1.575 1.513 1.491 1.471 1.409 1.452 1.555 1.415
Operating margin 0.243 0.210 0.182 0.190 0.179 0.186 0.195 0.159 0.165 0.169 0.160 0.173 0.140 0.094 0.011 0.032
ROA 0.102 0.093 0.079 0.078 0.066 0.057 0.054 0.043 0.042 0.037 0.031 0.019 −0.002 −0.043 −0.092 −0.124
CF/PPE 0.555 0.490 0.445 0.455 0.444 0.443 0.404 0.378 0.502 0.448 0.479 0.399 0.323 0.102 −0.195 −0.278
Capex/PPE 0.220 0.214 0.215 0.212 0.211 0.216 0.222 0.200 0.221 0.219 0.242 0.243 0.256 0.277 0.282 0.175
with smaller percentage differences in market leverage. BB+ firms have higher operating
margins but lower ROA and cash flow than BBB- firms. The investment rate of both BB+
and BBB- firms is around 22%.
X̄1 −X̄0
We also examine normalized differences in firm characteristics ∆X = √ 2 2
, where
S0 +S1

Sg2 is the sample variance of firm characteristic X in group g. Imbens and Wooldridge
(2009) advocate using the normalized difference as a scale-free measure of the difference in
distributions. They suggest that researchers should check that all normalized differences are
below 0.25 because linear regression methods can be sensitive to the specification otherwise.
At the investment-grade cutoff, all normalized differences are comfortably below the 0.25
threshold.24

D Flows are large relative to the investment of BB firms

The time series of aggregate flows into high-yield corporate bond mutual funds are from
the Investment Company Institute, the national association of U.S. investment companies.
At the end of 2008, the Investment Company Institute collected information on assets and
flows from 8,889 mutual funds with $9.6 trillion in assets under management. In our data,
high-yield mutual funds have $6 billion under management at the beginning of the sample
and $150 billion at the end.
The appropriate measure of flows should capture their magnitude relative to the capital
of firms close to the investment-grade cutoff, and also account for the time lag between fund
flows and bond issuance on one hand and issuance and investment on the other hand. To
accomplish these goals, we calculate cumulative flows over the four quarters [t − 4, t − 1] and
scale flows by the total PPE of firms rated BBB+ through BB-, P P Et−1 . Our results are
24
Panel B of Appendix Table XV reports normalized differences across adjacent credit ratings.

20
robust to calculating flows over other windows and using alternative scalings, in particular
scaling flows by total net assets (TNA) of high-yield mutual funds. Figure 1 shows the time-
series of high-yield fund flows relative to PPE and capital expenditures of firms rated BBB+
through BB-. Flows vary significantly over time and are large relative to the investment of
these firms. In our regressions, we standardize flows so that the coefficients can be interpreted
as the effect on investment of a one standard deviation increase in scaled flows.

IV Results

A Characteristics of matched BB+ and BBB- firms

Table II reports the characteristics of matched BB+ and BBB- firms. We match 2,818
out of 3,519 firm-quarter observations rated BB+ to 1,992 unique firm-quarter observations
rated BBB-. We report both the difference in means and the normalized difference.
Our matching procedure successfully picks BB+ and BBB- firms that have similar size.
Although in the full sample, BB+ firms are on average 27.8% smaller than BBB- firms, in
the matched sample, BB+ firms are actually slightly larger than matched BBB- firms. The
difference in size, however, is not statistically significant.
BB+ firms are still more levered than matched BBB- firms. The differences in book
and market leverage are around 0.05. BB+ firms also have lower Altman’s z-score and
interest coverage than matched BBB- firms. However, the normalized differences for these
measures of leverage are all below the 0.25 cutoff suggested by Imbens and Wooldridge (2009).
With the exception of ROA, none of the other differences are statistically or economically
significant. Overall, our matching procedure selects a sample of BB+ and BBB- firms that

21
 0.06 0.60

0.04 0.40

Flows/Capex
Flows/PPE

0.02
0.20

0.00
0.00

-0.02
-0.20
1986q1 1989q1 1992q1 1995q1 1998q1 2001q1 2004q1 2007q1

Flows/PPE Flows/Capex

Figure 1
High-Yield Fund Flows relative to PPE and Capex of BBB and BB Firms

This figure shows the time-series of flows into high-yield mutual funds. Monthly aggregate flows
into high-yield mutual funds are from the Investment Company Institute. Cumulative high-yield fund
flows calculated over four quarters are scaled by either total PPE or cumulative capital expenditures over
four quarters of CRSP/Compustat firms rated BBB+ through BB-, excluding financials and utilities. The
sample period is 1986Q1-2007Q4.

are similar along most dimensions.25 In our regressions, we are careful to control for the
remaining differences in observable characteristics such as leverage.
In addition to the full sample of BB+ firms, we consider two subsamples of firms that

25
In untabulated results, we do not find any differences in corporate governance, as measured by the G
Index (Gompers, Ishii, and Metrick, 2003) and its components, between the matched BB+ and BBB- firms.

22
 Table II
Characteristics of Matched BB+ and BBB- Firms

This table reports differences in the characteristics of matched BB+ and BBB- firms across three
alternative matched samples. The first one includes all BB+ firms that are matched to a BBB- firm. The
second one excludes the top 25% of BB+ firms by market leverage. The third one consists of BB+ firms
with positive outlooks. Each quarter a given firm rated BB+ is matched to a similar size firm in its industry
that is rated BBB-. Firms are considered to be of similar size if the ratio of their assets is within [0.5, 2].
Starting with the Fama-French 48 industries, we match firms using 38, 30, and 17 industry definitions until
we either have a successful match or are unable to match a given BB+ firm. In the full sample, 2,818
out of 3,519 firm-quarter observations rated BB+ are matched to 1,992 unique firm-quarter observations
rated BBB-. In the subsample of BB+ firms that excludes the top 25% by market leverage, 2,130 out of
2,645 firm-quarter observations rated BB+ are matched to 1,603 unique firm-quarter observations rated
BBB-. In the subsample of BB+ firms with positive outlook, 253 out of 417 firm-quarter observations
rated BB+ are matched to 247 unique firm-quarter observations rated BBB-. The sample period is
1986Q1-2007Q4. The table reports both the difference in means ∆X = X̄BB+ − X̄BBB− and the normalized
X̄BB+ −X̄BBB−
difference ∆X = √ 2 2
, where Sg2 is the sample variance of X for group g. Normalized difference is
SBB+ +SBBB−
a scale-free measure of the difference in distributions (Imbens and Wooldridge, 2009). Standard errors are
adjusted for clustering by firm. *, **, and *** denote statistical significance at 10%, 5%, and 1%.

Excluding top
25% of BB+ firms BB+ firms with
All BB+ firms by market leverage positive outlook
Normalized Normalized Normalized
Difference Difference Difference Difference Difference Difference
Assets 244 0.036 275 0.047 148 0.018
Book leverage 0.050∗∗ 0.147 0.015 0.042 0.055 0.136
Market leverage 0.053∗∗∗ 0.205 −0.006 −0.025 −0.002 −0.011
Z-score −0.111∗∗ −0.151 −0.098∗ −0.130 −0.010 −0.013
Interest coverage −1.348∗ −0.096 −0.875 −0.058 −1.114 −0.077
Cash/Assets −0.004 −0.038 0.001 0.005 −0.019 −0.143
PPE/Assets 0.022 0.065 0.012 0.034 0.029 0.080
Q −0.065 −0.066 0.014 0.013 0.135 0.116
Operating margin −0.001 −0.006 0.001 0.003 0.002 0.007
ROA −0.007∗∗ −0.070 0.000 0.003 0.017∗ 0.181
CF/PPE −0.044 −0.041 0.010 0.009 0.148 0.110
Capex/PPE −0.006 −0.025 0.006 0.025 0.030 0.112

are closer to the investment-grade cutoff. The first subsample excludes the top 25% of
BB+ firms by market leverage. The second subsample consists of BB+ firms with positive
outlooks.26 The differences in leverage are now greatly reduced. In the subsample that

26
In the subsample that excludes the top 25% of BB+ firms by market leverage, we match 2,130 BB+
observations to 1,603 unique BBB- firm-quarter observations. In the subsample of BB+ firms with positive

23
excludes the top 25% of BB+ firms by market leverage, the only remaining statistically
significant difference is in terms of the z-score.27 And in the subsample that consists of
BB+ firms with positive outlooks, the only statistically significant difference is in terms of
the ROA, with speculative-grade firms in fact being more profitable than investment-grade
ones. BB+ firms with positive outlooks also have higher values of Q and cash flow than
matched BBB- firms. Overall, the two subsamples select firms that are even closer to the
investment-grade cutoff than the ones in the full sample and result in an even better match.

B Flows increase the investment of BB+ firms relative to BBB-

firms

Table III reports the results of our baseline regressions. We regress the difference in
the investment rates of matched BB+ and BBB- firms on high-yield mutual fund flows and
differences in firm characteristics

CAP Xi,t CFi,t

∆ = α + βF lows · F lowst−1 + βQ · ∆Qi,t−1 + βCF · ∆ + εi,t (3)
P P Ei,t−1 P P Ei,t−1

where ∆X = X BB+ − X BBB− is the difference in firm characteristic X between matched

BB+ and BBB- firms. We use the procedure developed by Thompson (2006) to cluster the
standard errors by both firm and quarter.
Examining the results in column 1, the coefficient on flows is positive and statistically
significant. A one standard deviation increase in flows increases the investment of BB+

outlook, we match 253 BB+ observations to 247 unique BBB- first-quarter observations.
27
Of the four components of z-score, the only statistically significant difference is in retained earnings. There
are no statistically significance differences in profitability, sales, or working capital.

24
 Table III
Difference in the Investment Rates of Matched Firms and Fund Flows

This table reports the results of the regressions of the difference in the investment rates of matched
BB+ and BBB- firms on high-yield fund flows and differences in firm characteristics
CAP X �
∆ P P Ei,t−1
i,t
= α + βF lows · F lowst−1 + X βX · ∆Xi,t−1 + εi,t

where ∆X = X BB+ − X BBB− is the difference in firm characteristic X between matched BB+
and BBB- firms. Results for the full sample of matched BB+ and BBB- firms are in columns 1-3. Results
for the subsample of BB+ firms that excludes the top 25% of BB+ firms by market leverage are in columns
4-6. Results for the subsample of BB+ firms with positive outlooks are in columns 7-9. The sample period
is 1986Q1-2007Q4. Cumulative high-yield mutual fund flows over the four quarters [t − 4, t − 1] are scaled
by the total PPE of all firms rated BBB+ through BB-, P P Et−1 . The value of fund flows is standardized so
that the coefficient on flows represents the effect of one standard deviation change in fund flows. Standard
errors are adjusted for clustering by both firm and quarter using Thompson (2006). *, **, and *** denote
statistical significance at 10%, 5%, and 1%.

Excluding top
25% of BB+ firms BB+ firms
All BB+ firms by market leverage with positive outlook
(1) (2) (3) (4) (5) (6) (7) (8) (9)
∆Qi,t−1 0.044∗∗∗ 0.022 0.027∗∗∗ 0.045∗∗∗ 0.023 0.026∗∗ 0.026 0.001 0.004
(0.010) (0.014) (0.010) (0.010) (0.015) (0.011) (0.023) (0.032) (0.023)
CF
∆ P P Ei,t−1
i,t
0.074∗∗∗ 0.058∗∗∗ 0.071∗∗∗ 0.066∗∗∗ 0.056∗∗∗ 0.064∗∗∗ 0.043∗ −0.017 0.042∗
(0.014) (0.013) (0.014) (0.014) (0.013) (0.014) (0.024) (0.033) (0.025)
Flowst−1 0.014∗∗ 0.016∗∗ 0.014∗∗ 0.017∗∗ 0.017∗∗ 0.016∗∗ 0.012 0.013 0.019
(0.006) (0.007) (0.006) (0.008) (0.009) (0.008) (0.027) (0.031) (0.027)
∆Log(Assetsi,t−1 ) −0.050∗∗ −0.043∗∗ −0.048∗ −0.041∗ −0.104 −0.074
(0.023) (0.021) (0.027) (0.024) (0.077) (0.058)
∆Leveragei,t−1 −0.175∗∗∗−0.146∗∗∗ −0.177∗∗∗−0.191∗∗∗ −0.400∗∗ −0.331∗∗∗
(0.045) (0.038) (0.065) (0.048) (0.186) (0.116)
∆Z-scorei,t−1 0.011 0.021 −0.034
(0.014) (0.016) (0.047)
∆Coveragei,t−1 0.001 0.001 0.003
(0.001) (0.001) (0.002)
Constant −0.001 0.013 0.007 0.005 0.015 0.006 0.020 0.065∗∗ 0.026
(0.009) (0.010) (0.009) (0.010) (0.011) (0.010) (0.024) (0.033) (0.023)
N 2818 2147 2818 2130 1628 2130 253 172 253
Adjusted R2 0.138 0.144 0.153 0.122 0.130 0.140 0.048 0.093 0.094

firms relative to the investment of matched BBB- firms by 0.014, or about 7% of the mean
investment rate. The constant term is close to zero and not statistically significant, indicating
that on average matched firms have similar investment rates.

25
 Controlling for size, market leverage, z-score, and interest coverage in column 2 does
not affect the magnitude or the statistical significance of the coefficient on fund flows. The
number of observations drops by almost a quarter, primarily due to missing values of z-score.
Because neither z-score nor interest coverage have significant explanatory power, we omit
them from the regressions in column 3.
In columns 4-6, we estimate the same regressions as in columns 1-3 but using the sub-
sample of less levered BB+ firms. Although statistical significance is slightly weaker due to
smaller sample sizes, the economic magnitudes remain the same. In columns 7-9, we use the
sample of BB+ firms with positive outlooks. The sample size is about one-tenth of the full
sample, and the coefficient on flows is no longer statistically significant. The point estimates,
however, are remarkably similar.
Overall, Table III shows a statistically significant and economically meaningful effect of
fund flows on the investment of BB+ firms relative to similar firms rated BBB-.

C Results are robust to alternative matching procedures

Table IV shows that our results are robust to alternative matching procedures. Our
benchmark matching procedure successively matches BB+ and BBB- firms within Fama-
French 48, 38, 30, and 17 industries until we either have a successful match or are unable
to match a given BB+ firm. Columns 1-3 show that we get even stronger results when we
stop the match process at 30, 38, or 48 industries. In particular when we only match inside
48 industries in column 3, the coefficient on fund flows is 0.018, more than a quarter larger
than the coefficient of 0.014 in our benchmark specification.
In columns 4 and 5, we continue matching inside the 48 industries, but adjust the size
constraint. Specifically, in column 4 we relax the size constraint to require the ratio of book

26
 Table IV
Robustness to Alternative Matching Procedures

This table shows the robustness of our results to alternative matching procedures. Our benchmark
matching procedure starts with Fama-French 48 industries and matches firms within 38, 30, and 17 industry
definitions until we either have a successful match or are unable to match a given BB+ firm. In columns
1-3 we stop the match process at 30, 38, and 48 industries. In columns 4 and 5 we match within the 48
industries and require the asset ratio of matched firms to be within [1/3, 3] and [0.5, 1.5]. In columns 6-8
we once again match within 48, 38, 30, and 17 industries. In column 6, we match on assets and market
leverage, and require the asset ratio to be within [0.5, 2] and the absolute difference in leverage to be less
than 0.2. In column 7, we match on assets and z-score, and require the asset ratio to be within [0.5, 2] and
the absolute difference in z-score to be less than 0.6. In column 8, we match on assets and Q, and require
the asset ratio to be within [0.5, 2] and the absolute difference in Q to be less than 0.9. The sample period
is 1986Q1-2007Q4. Cumulative high-yield mutual fund flows over the four quarters [t − 4, t − 1] are scaled
by the total PPE of all firms rated BBB+ through BB-, P P Et−1 . The value of fund flows is standardized so
that the coefficient on flows represents the effect of one standard deviation change in fund flows. Standard
errors are adjusted for clustering by both firm and quarter using Thompson (2006). *, **, and *** denote
statistical significance at 10%, 5%, and 1%.

(1) (2) (3) (4) (5) (6) (7) (8)

∆Qi,t−1 0.026∗∗ 0.026∗∗ 0.032∗∗∗ 0.034∗∗∗ 0.028∗∗ 0.027∗∗ 0.024∗∗ 0.047∗∗
(0.010) (0.011) (0.012) (0.011) (0.014) (0.011) (0.010) (0.022)
CF
∆ P P Ei,t−1
i,t
0.082 ∗∗∗
0.070 ∗∗∗
0.070 ∗∗∗
0.065 ∗∗∗
0.074 ∗∗∗
0.065 ∗∗∗
0.049∗∗∗ 0.074∗∗∗
(0.015) (0.013) (0.014) (0.013) (0.016) (0.012) (0.012) (0.013)
Flowst−1 0.015∗∗ 0.017∗∗ 0.018∗∗ 0.016∗∗ 0.019∗∗ 0.014∗∗ 0.015∗∗ 0.011∗
(0.007) (0.007) (0.008) (0.007) (0.009) (0.007) (0.007) (0.007)
∆Log(Assetsi,t−1 ) −0.045∗∗ −0.049∗∗ −0.054∗∗ −0.030∗ −0.055∗ −0.026 −0.037∗∗ −0.044∗∗∗
(0.022) (0.022) (0.024) (0.016) (0.031) (0.018) (0.019) (0.016)
∆Leveragei,t−1 −0.147∗∗∗ −0.163∗∗∗ −0.164∗∗∗ −0.150∗∗∗ −0.179∗∗∗ −0.242∗∗∗ −0.168∗∗∗ −0.148∗∗∗
(0.041) (0.038) (0.043) (0.040) (0.046) (0.069) (0.045) (0.046)
Constant 0.008 0.008 0.010 0.013 0.007 0.005 0.007 0.013
(0.010) (0.010) (0.011) (0.010) (0.011) (0.009) (0.012) (0.010)
N 2437 2253 1890 2317 1605 2108 1917 2318
Adjusted R2 0.169 0.149 0.154 0.137 0.162 0.117 0.109 0.137
Match variables Assets Assets Assets Assets Assets Assets Assets Assets
Leverage Z-score Q
Fama-French 48, 38 48, 38 48 48 48 48, 38 48, 38
industry definitions 30 30, 17 30, 17 30, 17

assets to be within [1/3, 3]. The coefficient on fund flows is 0.016, somewhat smaller than in
column 3, but larger than in the benchmark specification of column 3 in Table III. In column
5 we tighten the size constraint to require the ratio of BB+ to matched BBB- firm assets to
be within [0.5, 1.5]. The effect of fund flows on investment is stronger for this sample.

27
 In columns 6-8, we match firms on assets and one of three other firm characteristics:
leverage, z-score, and Q.28 When matching on assets and leverage or assets and z-score, the
coefficient on fund flows is very similar to the benchmark specification. Matching on assets
and Q results in a coefficient on fund flows of 0.011 that is significant at better than 10%.

D Results are robust to alternative calculations of flows

Our results are also robust to alternative constructions of fund flows. In Panel A of
Table V, we calculate fund flows over different windows. In the first four columns, we
calculate flows over the four quarters [t − 3 − lag, t − lag]. In the last four columns, we
calculate flows over the two quarters [t−1−lag, t −lag]. Our results are robust to measuring
flows over two quarters. Consistent with there being a lag between flows and investment
(since it takes time to alter investment plans or issue bonds), the results are actually stronger
when using longer lags.
Nor do the results depend on how we scale fund flows. Panel B of Table V shows that we
get similar results when scaling flows by a) total PPE of all speculative-grade firms, b) total
assets of firms rated BBB+ through BB-, or c) total net assets of high-yield mutual funds.

E Rating, not continuous measures of credit quality, determines

the sensitivity to flows

Our results hold for a subsample of less levered BB+ firms for which there are virtually
no differences in observable characteristics between BB+ and BBB- firms. To further dispel
28
We use Fama-French 48, 38, 30, and 17 industries to look for the closest match according to the normalized
Euclidean distance. We require the ratio of assets to be within [0.5, 2] and the absolute differences in leverage,
z-score, and Q to be less than 0.2, 0.6, and 0.9 respectively. These values correspond to roughly one standard
deviation of each one of these three variables.

28
 Table V
Robustness to Alternative Calculations of Fund Flows

This table shows the robustness of our results to alternative calculations of fund flows. In Panel A,
we use alternative lags and windows for calculating fund flows. In columns 1-4, fund flows are calculated
over the four quarters [t − 3 − lag, t − lag]. In columns 5-8, fund flows are calculated over the two quarters
[t − 1 − lag, t − lag]. In Panel B, we calculate flows over the four quarters [t − 4, t − 1] and use alternative
scalings: a) total PPE of BBB+ through BB- firms, b) total PPE of speculative-grade firms, BB+ through
CCC-, c) fund total net assets, TNA, and d) total assets of BBB+ through BB- firms. The value of fund
flows is standardized so that the coefficient on flows represents the effect of one standard deviation change
in fund flows. The sample period is 1986Q1-2007Q4. Standard errors are adjusted for clustering by both
firm and quarter using Thompson (2006). *, **, and *** denote statistical significance at 10%, 5%, and 1%.

Panel A: Alternative Lags of Fund Flows

4 quarters of flows 2 quarters of flows
Lag 1 Lag 2 Lag 3 Lag 4 Lag 1 Lag 2 Lag 3 Lag 4
∆Qi,t−1 0.027∗∗∗ 0.026∗∗∗ 0.026∗∗∗ 0.027∗∗∗ 0.027∗∗∗ 0.026∗∗∗ 0.026∗∗∗ 0.026∗∗∗
(0.010) (0.010) (0.010) (0.010) (0.010) (0.010) (0.010) (0.010)
CF
∆ P P Ei,t−1
i,t
0.071∗∗∗ 0.071∗∗∗ 0.072∗∗∗ 0.072∗∗∗ 0.071∗∗∗ 0.071∗∗∗ 0.072∗∗∗ 0.072∗∗∗
(0.014) (0.014) (0.014) (0.015) (0.014) (0.014) (0.014) (0.014)
Flowst−1 0.014 ∗∗
0.015 ∗∗∗
0.015 ∗∗∗
0.014 ∗∗
0.012 ∗
0.015 ∗∗∗
0.015∗∗∗ 0.013∗∗
(0.006) (0.005) (0.005) (0.005) (0.007) (0.005) (0.005) (0.005)
∆Log(Assetsi,t−1 ) −0.043∗∗ −0.043∗∗ −0.044∗∗ −0.043∗∗ −0.044∗∗ −0.044∗∗ −0.044∗∗ −0.043∗∗
(0.021) (0.021) (0.021) (0.021) (0.021) (0.021) (0.021) (0.021)
∆Leveragei,t−1 −0.146∗∗∗ −0.148∗∗∗ −0.148∗∗∗ −0.147∗∗∗ −0.144∗∗∗ −0.147∗∗∗ −0.147∗∗∗ −0.147∗∗∗
(0.038) (0.038) (0.038) (0.038) (0.038) (0.038) (0.038) (0.038)
Constant 0.007 0.007 0.007 0.007 0.007 0.007 0.007 0.007
(0.009) (0.009) (0.009) (0.009) (0.009) (0.009) (0.009) (0.009)
N 2818 2818 2818 2818 2818 2818 2818 2818
Adjusted R2 0.153 0.153 0.153 0.153 0.152 0.153 0.153 0.152
Panel B: Alternative Scaling of Fund Flows
PPE of BBB+ PPE of BB+ Assets of BBB+
through BB- through CCC- TNA through BB-
∆Qi,t−1 0.027∗∗∗ 0.027∗∗∗ 0.027∗∗∗ 0.027∗∗∗
(0.010) (0.010) (0.010) (0.010)
CF
∆ P P Ei,t−1
i,t
0.071 ∗∗∗
0.071 ∗∗∗
0.071 ∗∗∗
0.071∗∗∗
(0.014) (0.014) (0.014) (0.015)
Flowst−1 0.014∗∗ 0.014∗∗ 0.014∗∗ 0.016∗∗∗
(0.006) (0.007) (0.005) (0.006)
∆Log(Assetsi,t−1 ) −0.043∗∗ −0.043∗∗ −0.043∗∗ −0.043∗∗
(0.021) (0.021) (0.021) (0.021)
∆Leveragei,t−1 −0.146∗∗∗ −0.146∗∗∗ −0.145∗∗∗ −0.146∗∗∗
(0.038) (0.038) (0.038) (0.038)
Constant 0.007 0.007 0.007 0.007
(0.009) (0.009) (0.009) (0.009)
N 2818 2818 2818 2818
Adjusted R2 0.153 0.153 0.153 0.154

29
any concern that our results might be driven by the investment of smaller and more levered
firms, we control for interactions of firm characteristics with fund flows. Table VI presents
the results. Column 1 is identical to column 2 in Table III, controlling for Q, cash flow,
size, leverage, z-score and interest coverage, but not their interactions with fund flows. As
we interact firm characteristics with fund flows in column 2, the direct effect of flows is
not affected. For compactness, we do not report the coefficients on the interactions of firm
characteristics with fund flows. Only the interaction of size with flows is consistently negative
and statistically significant.
In columns 3-6 we control for up to five powers of firm characteristics and their interac-
tions with fund flows. The direct effect of flows is not affected and actually gets stronger.
In particular, controlling for five powers of firm characteristics and their interactions with
flows increases the coefficient on flows from 0.016 in column 1 to 0.023 in column 6.

F Results are robust to controlling for macro variables

Our results so far indicate that the investment of BB+ firms is more sensitive to flows
into high-yield mutual funds than the investment of matched BBB- firms. Our identifying
assumption is that firms close to the investment-grade cutoff are subject to similar investment
opportunities shocks. If this assumption holds, the differential sensitivity of investment to
fund flows is evidence of the real effects of capital supply shocks in the presence of market
segmentation. Our empirical approach of matching on industry and size and controlling for
other firm characteristics, and their interactions with fund flows, is designed to ensure that
the identifying assumption holds so that our interpretation is valid.
However, there may still be concerns that investment opportunities are not quite the
same, and that fund flows are driven by the differential investment opportunities of lower-

30
 Table VI
Controlling for Interactions of Firm Characteristics with Fund Flows

This table reports the results of the regressions of the difference in the investment rates of matched
BB+ and BBB- firms on high-yield fund flows and differences in the characteristics of matched firms
interacted with fund flows
CAP X �
∆ P P Ei,t−1
i,t
= α + βF lows · F lowst−1 + X (βX · ∆Xi,t−1 + βX·F lows · ∆Xi,t−1 · F lowst−1 ) + εi,t

where ∆X = X BB+ − X BBB− is the difference in firm characteristic X between matched BB+
and BBB- firms. Coefficients on the interactions of firm characteristics with fund flows are not reported.
The sample period is 1986Q1-2007Q4. Cumulative high-yield mutual fund flows over the four quarters
[t − 4, t − 1] are scaled by the total PPE of all firms rated BBB+ through BB-, P P Et−1 . The value of fund
flows is standardized so that the coefficient on flows represents the effect of one standard deviation change
in fund flows. Standard errors are adjusted for clustering by both firm and quarter using Thompson
(2006). *, **, and *** denote statistical significance at 10%, 5%, and 1%.

(1) (2) (3) (4) (5) (6)

∆Qi,t−1 0.022 0.022∗ 0.019 0.017 0.018 0.025
(0.014) (0.013) (0.012) (0.015) (0.015) (0.016)
CF
∆ P P Ei,t−1
i,t
0.058∗∗∗ 0.058∗∗∗ 0.064∗∗∗ 0.077∗∗∗ 0.078∗∗∗ 0.091∗∗∗
(0.013) (0.013) (0.014) (0.015) (0.014) (0.019)
Flowst−1 0.016∗∗ 0.017∗∗ 0.019∗∗ 0.021∗∗ 0.022∗∗ 0.023∗∗
(0.007) (0.007) (0.009) (0.009) (0.010) (0.010)
∆Log(Assetsi,t−1 ) −0.050∗∗ −0.045∗∗ −0.043∗∗ −0.035 0.002 0.005
(0.023) (0.022) (0.022) (0.026) (0.037) (0.047)
∆Leveragei,t−1 −0.175∗∗∗ −0.168∗∗∗ −0.196∗∗∗ −0.184∗∗ −0.180∗∗ −0.189∗∗
(0.045) (0.046) (0.054) (0.074) (0.075) (0.090)
∆Z-scorei,t−1 0.011 0.011 0.008 −0.009 −0.009 −0.021
(0.014) (0.014) (0.014) (0.018) (0.019) (0.021)
∆Coveragei,t−1 0.001 0.001 0.001∗ 0.003∗∗∗ 0.002∗∗ 0.002
(0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
Constant 0.013 0.014 0.019∗ 0.018∗ 0.011 0.009
(0.010) (0.010) (0.010) (0.011) (0.013) (0.013)
N 2147 2147 2147 2147 2147 2147
Adjusted R2 0.144 0.152 0.166 0.170 0.172 0.171
Interactions with flows No Yes Yes Yes Yes Yes
Powers of characteristics 1 1 1-2 1-3 1-4 1-5

rated firms. A natural alternative interpretation is that lower-rated firms are more sensitive
to the business cycle and that fund flows are picking up this greater sensitivity. This would
be the case if the investment opportunities of lower-rated firms are procyclical and high yield
fund flows are proxying for the state of the macroeconomy.

31
 We address this possibility by directly controlling for a number of macroeconomic vari-
ables.29 To make the test as stringent as possible, we focus on the subsample of less levered
BB+ firms and control for size, leverage, z-score, and interest coverage, as well as their in-
teractions with fund flows. The variables we control for are the level of the VIX, the term
spread, the Baa-Aaa credit spread, the aggregate stock market return, and GDP growth.
We measure these variables as of quarter t − 1, but our results are robust to using average
values over the four quarters [t − 4, t − 1] or to using contemporaneous values.
Table VII presents the results. The first column shows the basic results in this specifica-
tion without controlling for any macro variables. The next 5 columns control for each macro
variable individually. Only the term spread is significant at 10%, and the coefficients on flows
are significant and of similar magnitudes to our previous results. The final column controls
for all of our macro variables simultaneously. Again, the coefficient on flows is unaffected.
Thus, it seems unlikely that the differential sensitivity of BB+ investment to fund flows is
driven by macroeconomic factors.

G No differential sensitivity to flows around other cutoffs

Next we conduct falsification tests using matched firm pairs around other rating cutoffs.
If our results were driven by the differential investment opportunities of lower-rated firms,
we might expect lower-rated firms to have a higher sensitivity of investment to fund flows
than higher-rated firms around every rating cutoff. To test this hypothesis, for each credit
rating cutoff from A through B we match firms just below the cutoff with firms just above

29
Our results are also robust to excluding the two recessions in our sample period, as well as to excluding
any given year.

32
 Table VII
Controlling for Macroeconomic Variables

This table reports the results of the regressions of the difference in the investment rates of matched
BB+ and BBB- firms on high-yield fund flows, differences in firm characteristics interacted with high-yield
fund flows, and macroeconomic variables
CAP Xi,t �
∆ = α + βF lows · F lowst−1 + (βX · ∆Xi,t−1 + β X · F lows · ∆Xi,t−1 · F lowst−1 ) + εi,t
P P Ei,t−1
X

where ∆X = X BB+ − X BBB− is the difference in firm characteristic X between matched BB+ and BBB-
firms.. Macroeconomic variables are defined in Appendix Table XIII, and are standardized so that the
coefficients represent the effect of one standard deviation change in the explanatory variables. The sample of
matched BB+ and BBB- firms excludes the top 25% of BB+ firms by market leverage. The sample period
is 1986Q1-2007Q4. Standard errors are adjusted for clustering by both firm and quarter using Thompson
(2006). *, **, and *** denote statistical significance at 10%, 5%, and 1%.

(1) (2) (3) (4) (5) (6) (7)

∆Qi,t−1 0.023 0.021 0.022 0.022 0.023 0.023 0.021
(0.014) (0.014) (0.014) (0.014) (0.014) (0.014) (0.014)
CF
∆ P P Ei,t−1
i,t
0.056∗∗∗ 0.057∗∗∗ 0.056∗∗∗ 0.056∗∗∗ 0.056∗∗∗ 0.056∗∗∗ 0.057∗∗∗
(0.013) (0.013) (0.013) (0.013) (0.013) (0.013) (0.013)
Flowst−1 0.019 ∗∗
0.016 ∗
0.022 ∗∗
0.019 ∗∗
0.018 ∗∗
0.018∗∗ 0.018∗∗
(0.009) (0.008) (0.009) (0.009) (0.009) (0.008) (0.009)
∆Log(Assetsi,t−1 ) −0.042∗ −0.043∗ −0.043∗ −0.042∗ −0.042∗ −0.042∗ −0.045∗
(0.025) (0.026) (0.025) (0.025) (0.025) (0.025) (0.026)
∆Leveragei,t−1 −0.170∗∗ −0.177∗∗∗ −0.169∗∗∗ −0.170∗∗∗ −0.169∗∗ −0.170∗∗ −0.176∗∗∗
(0.066) (0.066) (0.065) (0.066) (0.066) (0.066) (0.065)
∆Z-scorei,t−1 0.022 0.022 0.023 0.022 0.022 0.022 0.024
(0.017) (0.017) (0.017) (0.017) (0.017) (0.017) (0.017)
∆Coveragei,t−1 0.001 0.001 0.001 0.001 0.001 0.001 0.001
(0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
VIXt−1 0.004 0.007
(0.010) (0.011)
Term spreadt−1 −0.015∗ −0.015∗
(0.008) (0.008)
Credit spreadt−1 −0.001 −0.002
(0.010) (0.011)
Stock market returnt−1 0.004 0.004
(0.005) (0.005)
GDP growtht−1 0.001 0.001
(0.007) (0.007)
Constant 0.015 0.014 0.015 0.015 0.015 0.015 0.014
(0.011) (0.011) (0.010) (0.011) (0.011) (0.011) (0.010)
N 1628 1613 1628 1628 1628 1628 1613
Adjusted R2 0.138 0.138 0.141 0.138 0.138 0.138 0.141

33
the cutoff that are of similar size and in the same industry.30 For example, we match firms
rated A with firms rated A+. As there are few firms rated above A or below B, we do not
report the results for cutoffs above A and below B.31 To focus on firms closer to each cutoff
and get a better match on both size and leverage, we exclude the top 25% of most levered
firms below the cutoff.
Each column of Table VIII reports the results of our placebo regressions for firms with
the credit rating specified by column heading and matched to firms rated one notch higher.
Thus, the first column compares the investment of A firms with the investment of matched
A+ firms. Our regressions control for size, leverage, z-score, and interest coverage, as well
as the interactions of firm characteristics with fund flows.
Other than the BB+ vs. BBB- cutoff, the only other cutoff for which there is a statistically
significant effect of fund flows on investment is BBB+ vs. A-. The effect goes the other way—
the investment of BBB+ firms is less sensitive to fund flows than the investment of matched
A- firms. Appendix Table XVIII indicates that our match of BBB+ and A- firms is far from
perfect. There are still large differences in Q and interest coverage, which may explain the
results we find.
Our results indicate that the investment-grade cutoff is the only one where the investment
of firms below the cutoff is more sensitive to fund flows than the investment of firms above
the cutoff. Importantly, the lack of statistically significant results around other cutoffs is
not due to smaller sample sizes and weaker power. Other than the A- vs. A cutoff, all other

30
As before, we successively match firms within Fama-French 48, 38, 30, and 17 industries until we either
have a successful match or are unable to match a given firm. We consider firms to be of similar size if the
ratio of their book assets is within [0.5, 2].
31
We find similar results, i.e., that investment of lower rated firms is not more sensitive to fund flows than
investment of higher rated firms, for these other cutoffs.

34
 Table VIII
Placebo Regressions

This table reports the results of the placebo regressions of the difference in the investment rates of matched firms around
other credit rating cutoffs on fund flows and differences in the characteristics of matched firms interacted with fund flows
CAP X i,t
∆ P P Ei,t−1 = α + βF lows · F lowst−1 + X (βX · ∆Xi,t−1 + βX·F lows · ∆Xi,t−1 · F lowst−1 ) + εi,t
�

where ∆X = X R − X R̂ is the difference in firm characteristic X between firms rated R and matched firms rated one notch
higher, R̂. In each column the sample consists of firms specified by the column name and matched firms rated one notch higher.
For example, the sample in the first column is firms rated A and matched firms rated A+. Around each rating cutoff, top 25%
of the most levered firms below the cutoff are excluded. Characteristics of matched firms are reported in Appendix Table XVIII.
The interactions of firm characteristics with fund flows are included but not reported. The sample period is 1986Q1-2007Q4.
Cumulative high-yield mutual fund flows over the four quarters [t − 4, t − 1] are scaled by the total PPE of all firms rated BBB+
through BB-, P P Et−1 . The value of fund flows is standardized so that the coefficient on flows represents the effect of one standard
deviation change in fund flows. Standard errors are adjusted for clustering by both firm and quarter using Thompson (2006). *,
**, and *** denote statistical significance at 10%, 5%, and 1%.

A A- BBB+ BBB BBB- BB+ BB BB- B+ B

35
∆Qi,t−1 0.001 0.007 0.028∗∗∗ 0.026∗∗ 0.043∗∗∗ 0.023 0.021 0.051∗∗∗ 0.042∗∗∗ 0.064∗∗∗
(0.007) (0.011) (0.006) (0.011) (0.009) (0.014) (0.013) (0.012) (0.011) (0.012)
CF i,t
∆ P P Ei,t−1 0.061∗∗∗ 0.037∗∗∗ 0.055∗∗∗ 0.041∗∗ 0.071∗∗∗ 0.056∗∗∗ 0.046∗∗∗ 0.042∗∗∗ 0.034∗∗∗ 0.024∗∗∗
(0.017) (0.010) (0.015) (0.016) (0.011) (0.013) (0.010) (0.010) (0.007) (0.008)
Flowst−1 0.003 0.008 −0.020∗∗∗ 0.005 0.001 0.019∗∗ −0.005 −0.005 0.005 0.008
(0.007) (0.007) (0.008) (0.005) (0.005) (0.009) (0.009) (0.008) (0.008) (0.011)
∆Log(Assetsi,t−1 ) −0.006 −0.024 −0.012 −0.043∗∗ −0.028 −0.042∗ −0.012 −0.064∗∗∗ −0.075∗∗∗ −0.037
(0.018) (0.024) (0.022) (0.020) (0.019) (0.025) (0.022) (0.020) (0.021) (0.030)
∆Leveragei,t−1 −0.296∗∗∗ −0.243∗∗ −0.026 −0.165∗∗∗ −0.039 −0.170∗∗ −0.225∗∗∗ −0.156∗∗∗ −0.255∗∗∗ −0.219∗∗∗
(0.079) (0.118) (0.073) (0.049) (0.043) (0.066) (0.052) (0.038) (0.035) (0.045)
∆Z-scorei,t−1 −0.045∗∗∗ −0.009 0.005 0.029 0.018 0.022 0.027∗ 0.025∗ 0.016 0.047∗∗∗
(0.015) (0.018) (0.017) (0.018) (0.014) (0.017) (0.015) (0.013) (0.012) (0.015)
∆Coveragei,t−1 0.001∗∗∗ −0.000 0.000 −0.001 0.000 0.001 0.001 0.001∗∗ 0.001∗ −0.000
(0.000) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
Constant −0.002 −0.003 0.023∗∗∗ 0.000 0.008 0.015 0.028∗∗∗ 0.007 0.028∗∗∗ 0.044∗∗∗
(0.008) (0.011) (0.009) (0.008) (0.007) (0.011) (0.010) (0.008) (0.009) (0.013)
N 1777 1510 1643 2625 2480 1628 2472 4156 4093 1966
Adjusted R2 0.154 0.083 0.137 0.111 0.146 0.138 0.151 0.176 0.167 0.154
cutoffs have more observations than our sample of BB+ and BBB- firms.

H Comovement of investment within credit grade

The last three sections suggest that our results are not driven by fund flows picking
up the differential investment opportunities of BB+ firms. Instead, flows into high-yield
mutual funds are likely driven by the performance and investment opportunities of BB-,
B+, and B firms, which together account for 61% of the total number of speculative-grade
firms. In comparison, the share of BB+ firms is only 11.6%. Thus, the effect of fund flows
on the investment of BB+ firms may be a spillover effect due to categorization. If investors
emphasize differences between broad categories, such as investment- versus speculative-grade
firms, and overlook differences within these broad categories, they may treat two firms in the
same category similarly, even if the firms have very different performance and investment
prospects (Mullainathan, 2000; Barberis and Shleifer, 2003).32 Categorization may thus
cause firms at the boundary to behave more like dissimilar firms in the rest of the category
and less like fundamentally similar firms just across the boundary.
Table IX provides evidence consistent with this idea by examining whether the invest-
ment of BB+ firms comoves with the investment of unrelated speculative-grade firms further
away from the cutoff. We define unrelated firms as firms outside firm i’s industry rated A
through BBB+ for investment grade and BB- through B for speculative grade. These rating
restrictions on unrelated speculative-grade firms are intended to ensure that they are in fact
unrelated and that they do not face immediate financial distress. To accomplish the first
objective we omit firms rated BB to create a buffer between BB+ and unrelated speculative-
32
For example, the difference in 5-year default rates between investment- and speculative-grade firms is
14%, but the difference in default rates between the lowest (CCC through C) and highest rated (BB+)
speculative-grade firms is 39%, almost three times as much.

36
grade firms. As a result, the BB+ firms are quite different from the rest of the category. For
example, while average 5-year default rates for BBB- and BB+ firms are 3.74% and 5.41%,
average 5-year default rates for BB-, B+, and B firms are 12.32%, 17.65%, and 23.84%. To
accomplish the second objective we omit firms rated B- and below. These firms are likely to
face immediate financial distress. In particular, B- firms have very high leverage, negative
profitability in terms of ROA and operating cash flow, and even negative interest cover-
age. We impose symmetric restrictions for unrelated investment-grade firms by omitting one
notch (BBB) and using the next three notches: BBB+, A-, and A.
The columns of Table IX alternatively use Fama-French 48, 38, and 30 industries to define
unrelated firms. For each industry definition, we regress the difference in the investment
rates of matched firms on either (i) unrelated speculative-grade investment and unrelated
investment-grade investment, or (ii) unrelated speculative-grade investment alone. We use
both specifications because the two investment series are highly correlated. The results in
Table IX show that unrelated speculative-grade investment drives out unrelated investment-
grade investment, but the presence of unrelated investment-grade investment as a regressor
increases the standard errors. When we regress the difference in the investment rates of
matched firms on unrelated speculative-grade investment alone, the coefficient is unchanged
and is significant at the 5% or 10% level depending on industry definition.33 The economic
magnitude is relatively large. If investment of firms rated BB- through B that are outside firm
i’s Fama-French 48 industry increases by one standard deviation (about 0.040), investment
of firm i increases by about 0.013 relative to the investment of matched BBB- firm.

33
In comparison, when we regress the difference in the investment rates of matched firms on unrelated
investment-grade investment alone, we get similar but statistically insignificant coefficients.

37
 Table IX
Comovement of Investment within Credit Grade

This table reports the results of the regressions of the difference in the investment rates of matched
BB+ and BBB- firms on the investment rates of unrelated investment- and speculative-grade firms and
differences in the characteristics of matched firms
CAP X �
∆ P P Ei,t−1
i,t
= α + βSGrade · Speculativei,t + βIGrade · Investmenti,t + X βX · ∆Xi,t−1 + εi,t

where ∆X = X BB+ − X BBB− is the difference in firm characteristic X between matched BB+
and BBB- firms. Average investment of unrelated firms is calculated using firms outside firm i’s industry,
using Fama-French 48, 38, or 30 industry definitions. Average investment of unrelated investment-grade
firms is calculated using firms rated A through BBB+. Average investment of unrelated speculative-grade
firms is calculated using firms rated BB- through B. The sample period is 1986Q1-2007Q4. Standard
errors are adjusted for clustering by both firm and quarter using Thompson (2006). *, **, and *** denote
statistical significance at 10%, 5%, and 1%.

Fama-French 48 Fama-French 38 Fama-French 30

(1) (2) (3) (4) (5) (6)
∆Qi,t−1 0.027∗∗∗ 0.027∗∗∗ 0.027∗∗∗ 0.027∗∗∗ 0.027∗∗∗ 0.027∗∗∗
(0.010) (0.010) (0.010) (0.010) (0.010) (0.010)
CF
∆ P P Ei,t−1
i,t
0.072 ∗∗∗
0.072∗∗∗ 0.072 ∗∗∗
0.072∗∗∗ 0.072 ∗∗∗
0.072∗∗∗
(0.015) (0.015) (0.015) (0.015) (0.015) (0.015)
Unrelated Investment Gradei,t −0.089 −0.051 −0.054
(0.265) (0.258) (0.262)
Unrelated Speculative Gradei,t 0.382∗ 0.335∗∗ 0.325 0.298∗ 0.346∗ 0.318∗∗
(0.211) (0.161) (0.213) (0.164) (0.207) (0.158)
∆Log(Assetsi,t−1 ) −0.044∗∗ −0.044∗∗ −0.044∗∗ −0.044∗∗ −0.044∗∗ −0.044∗∗
(0.021) (0.021) (0.021) (0.021) (0.021) (0.021)
∆Leveragei,t−1 −0.144∗∗∗ −0.144∗∗∗ −0.144∗∗∗ −0.144∗∗∗ −0.145∗∗∗ −0.145∗∗∗
(0.038) (0.038) (0.038) (0.038) (0.038) (0.038)
Constant −0.069 −0.076∗ −0.063 −0.067∗ −0.067 −0.072∗
(0.045) (0.039) (0.045) (0.040) (0.044) (0.039)
N 2818 2818 2818 2818 2818 2818
Adjusted R2 0.153 0.153 0.152 0.152 0.152 0.153

I Higher sensitivity of firms without access to other financing

sources

Are certain types of firms more sensitive to fund flows? In principle financially constrained
firms with limited access to other sources of financing should be more sensitive to fund flows.
Small firms and firms that do not pay dividends are more likely to be financially constrained

38
(Fazzari, Hubbard, and Petersen, 1988; Baker, Stein, and Wurgler, 2003). Similarly, firms
with low cash flow from operations and high dependence on external financing should be
more sensitive to fund flows (Rajan and Zingales, 1998). On the other hand, the investment
of firms that can borrow from banks or have access to the asset-backed securities market
should be less sensitive to fund flows.
Table X tests this idea by estimating our investment regressions for six different sample
splits. In columns 1 and 2, we split BB+ firms by size. For smaller firms, the coefficient on
flows is positive and statistically significant. For larger firms, the coefficient is positive but
not statistically significant. Although we cannot reject that the two coefficients are the same,
it is encouraging that the effect of fund flows is stronger in the sample of smaller BB+ firms.
Note that our methodology is somewhat different than the typical cross-sectional analysis.
We are keeping matched pairs together, but splitting the sample of matched pairs based on
BB+ firm characteristics. Thus our results imply that small BB+ firms are more sensitive to
flows relative to their small BBB- matches than large BB+ firms are relative to their large
BBB- matches.34
In columns 3 and 4, we split BB+ firms by whether they pay dividends. For dividend
paying firms, there is no differential sensitivity of investment to fund flows between BB+ and
BBB- firms. The investment of BB+ firms that do not pay dividends, on the other hand, is
strongly sensitive to fund flows. The coefficient on flows for these firms is 0.027, more than
five times as large as for dividend payers.
We next split BB+ firms by their cash flow from operations in columns 5 and 6, and
the Rajan and Zingales (1998) measure of external dependence in columns 7 and 8. The
34
Match quality is similar across the different sample splits, and we are careful to include the interactions of
firm characteristics with fund flows to make sure that our cross-sectional results are not diven by differences
in match quality.

39
 Table X
Cross-Sectional Splits

This table reports the results of the regressions of the difference in the investment rates of matched BB+ and BBB- firms
on high-yield fund flows and differences in the characteristics of matched firms interacted with fund flows
CAP X i,t
∆ P P Ei,t−1 = α + βF lows · F lowst−1 + X (βX · ∆Xi,t−1 + βX·F lows · ∆Xi,t−1 · F lowst−1 ) + εi,t
�

estimated separately for subsamples of BB+ firms split by firm size, dividends, cash flows, external dependence, having a
bank loan rating, and access to the asset-backed securities market. ∆X = X BB+ − X BBB− is the difference in firm characteristic
X between matched BB+ and BBB- firms. Small firms are the ones below the median of assets for BB+ firms. Dividend payers
are firms with positive cash dividends. Low cash flow firms are the ones below the median of the cash-to-assets ratio for BB+
firms. Appendix Table XX reports for each Fama-French 48 industry the value of the Rajan and Zingales (1998) measure of
external dependence. Industry external dependence is calculated using data for the 1970-1985 period. Bank loan rating indicates the
existence of a bank loan rating. Top 5 industries by the share of asset- and mortage-backed securities issuance in total industry-level
bond issuance are Electrical Equipment, Personal Services, Automobiles and Trucks, Machinery, and Retail. Appendix Table XXI
reports for each Fama-French 48 industry the share of asset- and mortgage-backed issuance in total bond issuance by publicly traded
non-financial domestic corporations. The sample period is 1986Q1-2007Q4 in all regressions except for the bank loan rating split,
where the sample period is 1986Q1-2005Q2. Standard errors are adjusted for clustering by both firm and quarter using Thompson

40
(2006). *, **, and *** denote statistical significance at 10%, 5%, and 1%.
Top 20 External Bank Top 5
Small Size Dividend Payer Low Cash Flow Dependence Industry Loan Rating ABS Industry
Yes No Yes No Yes No Yes No Yes No Yes No
∆Qi,t−1 0.024 0.032∗∗∗0.036∗∗∗ 0.015 0.048∗∗∗ 0.020 0.039∗∗∗ 0.002 0.010 0.066∗∗∗0.018 0.028∗∗
(0.015) (0.012) (0.012) (0.013) (0.014) (0.013) (0.011) (0.018) (0.011) (0.017) (0.012) (0.012)
CFi,t
∆ P P Ei,t−1 0.086∗∗∗0.058∗∗∗0.102∗∗∗ 0.046∗∗∗ 0.086∗∗∗ 0.065∗∗∗ 0.053∗∗∗ 0.084∗∗∗ 0.088∗∗∗0.054∗∗∗0.063∗∗∗0.073∗∗∗
(0.019) (0.019) (0.018) (0.018) (0.024) (0.017) (0.018) (0.019) (0.019) (0.018) (0.020) (0.016)
Flowst−1 0.017∗∗ 0.013 0.005 0.027∗∗ 0.023∗∗∗ 0.009 0.032∗∗∗ 0.002 0.008 0.034∗∗∗ −0.006 0.018∗∗
(0.009) (0.010) (0.006) (0.012) (0.008) (0.008) (0.008) (0.007) (0.008) (0.013) (0.011) (0.008)
∆Log(Assetsi,t−1 ) −0.035 −0.051∗ −0.027 −0.061∗ −0.060∗∗ −0.022 −0.074∗∗ −0.007 −0.046∗ −0.039 −0.087∗∗∗ −0.033
(0.026) (0.030) (0.021) (0.032) (0.027) (0.024) (0.032) (0.025) (0.025) (0.048) (0.030) (0.022)
∆Leveragei,t−1 −0.096 −0.193∗∗∗−0.108∗∗ −0.177∗∗∗−0.145∗∗∗−0.096 −0.185∗∗∗ −0.097 −0.153∗∗∗−0.183∗∗−0.255∗∗∗−0.108∗∗
(0.060) (0.039) (0.048) (0.052) (0.043) (0.072) (0.046) (0.068) (0.048) (0.077) (0.076) (0.047)
Constant 0.001 0.009 −0.004 0.019 0.011 0.014 0.010 0.002 0.008 0.017 0.023 0.003
(0.013) (0.012) (0.009) (0.015) (0.012) (0.011) (0.013) (0.011) (0.012) (0.022) (0.015) (0.011)
N 1409 1409 1576 1234 1409 1409 1493 1325 1773 584 583 2235
Adjusted R2 0.164 0.159 0.217 0.122 0.165 0.125 0.170 0.183 0.171 0.204 0.220 0.154
latter is meant to capture at the industry level the share of capital expenditures that is
financed externally versus using internal cash flow. Appendix Table XX reports the external
dependence measure for each of the Fama-French 48 industries.35 We find that the investment
of BB+ firms with low cash flow and BB+ firms in the top twenty industries by external
dependence is sensitive to fund flows.36 The investment of other firms, in particular ones
in industries with low external dependence, is not at all affected by fund flows. This cross-
sectional split is particularly informative, since a firm’s industry cannot reveal anything
about its credit quality (relative to a matched firm in the same industry). Thus, the fact
that we find a pronounced differential sensitivity of investment to fund flows for firms in
industries with high external dependence is strong evidence that financial constraints are at
work.
Next, we measure firms’ ability to borrow from banks by whether they have a bank loan
rating. Such firms should find it easier to substitute to the syndicated loan market when
fund flows are low. This is what we find in columns 9 and 10. The investment of firms with
a loan rating is not sensitive to fund flows, but the investment of firms without a loan rating
is very sensitive to fund flows. A one standard deviation increase in fund flows increases the
investment of these firms by 0.034, or about 15% of their mean level of investment.
Finally, in columns 11 and 12 we split firms by whether they have access to the asset-
backed securities market. Issuing asset- or mortgage-backed securities through a bankruptcy-
remote trust can allow firms to tap investment-grade sources of financing. Since certain

35
To calculate industry external dependence we use nonoverlapping data from the 1970-1985 period to
prevent the realizations of fund flows during our sample period from affecting our measures of external
dependence. We get similar results, however, when using the same sample period to measure external
dependence and estimate our investment regressions.
36
We get similar results when splitting firms into top ten and top fifteen industries.

41
assets are much easier to securitize than others, access to the asset-backed securities market
depends on the nature of firm assets. For instance, credit cards, car loans, and certain
types of machinery and equipment are easier to borrow against in the asset-backed securities
market. We therefore measure firms’ ability to substitute to the asset-backed market by
whether they are in an industry with significant issuance of asset- and mortgage-backed
securities. Appendix Table XXI reports for each Fama-French 48 industry the share of
asset- and mortgage-backed securities in total bond issuance over our sample period by non-
financial firms. Excluding the Other industry, the top five industries by ABS issuance are
Electrical Equipment, Personal Services, Automobiles and Trucks, Machinery, and Retail.37
With the exception of Construction, asset- and mortgage-backed securities account for less
than 10% of the total bond issuance for firms outside the top five industries by ABS issuance.
This suggests limited reliance on and ability to issue asset-backed securities by firms outside
the top five industries. We therefore label firms as having access to the asset-backed market
if they are in the top five industries by asset-backed issuance.38
The results in columns 11 and 12 indicate that the investment of firms in the top five
ABS industries is not sensitive to fund flows. In fact the coefficient on fund flows is negative
but not statistically significant. It is the investment of firms in other industries that responds
to fund flows.
Taken together, our results in Table X indicate that the investment of firms with limited
ability to substitute away from the high-yield market responds strongly to flows into high-
yield mutual funds, while the investment of other firms is generally not affected.
37
Two firms—General Environmental Management and various ARG Funding trusts of AutoNation—
account for almost all asset- and mortage-backed issuance of the Other industry. Our results are robust
to defining the top five industries by ABS issuance to include Other and exclude Retail.
38
We get qualitatively similar though weaker results when we look at firms in the top ten industries.

42
J BB+ bond issuance is more sensitive to flows than BBB- bond

issuance

So far we have explored the connection between fund flows and firm investment without
documenting a particular mechanism. In principle, our results could be driven by an increase
in the availability of financing or a decrease in the cost of capital. In the first explanation,
investment and issuance are tightly linked. Flows into high-yield mutual funds increase
the supply of capital invested in high-yield bonds. This allows speculative-grade firms to
increase their bond issuance and use the proceeds to increase their investment. In the
second explanation, issuance is not necessarily involved. Fund flows lower bond yields and
the required rates of return perceived by speculative-grade firms. These firms then lower
their hurdle rates and invest more, potentially financing this investment from sources other
than the high-yield bond market.
We are interested in understanding the mechanism that drives our results. In the absence
of high-quality firm-level data on bond prices, documenting cost of capital effects will be
difficult. However, we can document the effect of fund flows on bond issuance. We continue to
use the same empirical methodology, regressing the difference in the bond issuance of matched
BB+ and BBB- firms on high-yield mutual fund flows and differences in firm characteristics.
Table XI presents the results. In the first three columns, we scale issuance by firm assets
and fund flows by the total assets of all firms rated BBB+ through BB-. The first column
shows that BB+ issuance is more sensitive to fund flows that BBB- issuance. A one-standard
deviation increase in flows increases issuance (as a fraction of assets) by 0.40%. Our power is
somewhat limited here since there are only 179 BB+ issuance events and 189 BBB- issuance
events in our data. However, the coefficient on flows still significant at the 10% level. The
second column shows the results are robust to controlling for size, Q, and leverage. The

43
third column shows that we obtain substantially stronger results if we simply use a dummy
for positive fund flows, instead of the continuous variable.
Columns 4-6 repeat the exercise normalizing issuance by firm P P E and flows by the total
P P E of all firms rated BBB+ through BB-. The results here are a bit weaker statistically
because low values of P P E for a few firms create outliers in normalized issuance. However,
the coefficient on flows is positive in columns 4 and 5, and the coefficient on the positive
flows dummy is positive and significant in column 6. Overall, the results strongly suggest
that the bond issuance of BB+ firms increases relative to that of their BBB- matches when
fund flows are high.
Finally we examine cross-sectional differences in the sensitivity of issuance to fund flows.
As before, we keep matched pairs together and split the sample of matched pairs based on
the characteristics of the BB+ firm. Since our power is relatively low even before splitting
the sample, we use our strongest specification from Table XI, scaling issuance by assets and
using a dummy for positive fund flows.
Table XII presents the results. Columns 1 and 2 show that the coefficient on flows is
larger and more significant for small BB+ firms relative to their small BBB- matches than
it is for large BB+ firms. However, the coefficients for small firms and large firms are of
similar magnitudes and their difference is not statistically significant. Columns 3 and 4 show
that non-dividend payers have a stronger differential sensitivity of issuance to fund flows
than dividend payers. In columns 5-8, we see that the sensitivity of issuance to fund flows is
roughly the same for low cashflow and high cashflow firms and for firms with low and high
external dependence. In columns 9 and 10, the coefficient on flows is significant for firms
that do have a bank loan rating, but insignificant for those that do not. This is somewhat
surprising, but the coefficients are roughly the same magnitude and there are relatively few

44
 Table XI
Bond Issuance by Matched Firms and Fund Flows

This table reports the results of the regressions of the difference in bond issuance by matched BB+
and BBB- firms on high-yield fund flows and differences in firm characteristics
�
∆Issuancei,t = α + βF lows · F lowst−1 + X βX · ∆Xi,t−1 + εi,t

where ∆X = X BB+ − X BBB− is the difference in firm characteristic X between matched BB+
and BBB- firms. Issuance of non-convertible, not asset- or mortgage-backed bonds from SDC is scaled by
lagged assets in columns 1-3 and by lagged PPE in columns 4-6. Cumulative high-yield mutual fund flows
over the four quarters [t − 4, t − 1] are scaled by the total assets of all firms rated BBB+ through BB-,
Assetst−1 , in columns 1-3, and by the total PPE of all firms rated BBB+ through BB-, P P Et−1 . The value
of fund flows is standardized so that the coefficient on flows represents the effect of one standard deviation
change in fund flows. The sample period is 1986Q1-2007Q4. Standard errors are adjusted for clustering
by both firm and quarter using Thompson (2006). *, **, and *** denote statistical significance at 10%, 5%,
and 1%.

Issuance and fund flows scaled by

Assets PPE
(1) (2) (3) (4) (5) (6)
Flowst−1 0.004∗ 0.004∗ 0.033 0.033
(0.002) (0.002) (0.021) (0.021)
Positive flowst−1 0.008∗∗∗ 0.072∗∗
(0.003) (0.032)
∆Qi,t−1 −0.001 −0.001 0.004 0.005
(0.001) (0.001) (0.014) (0.015)
∆Log(Assetsi,t−1 ) −0.012∗∗∗ −0.012∗∗∗ −0.036 −0.037
(0.004) (0.004) (0.064) (0.064)
∆Leveragei,t−1 0.001 0.001 0.141 0.138
(0.007) (0.007) (0.086) (0.086)
Constant 0.001 0.002 −0.002 −0.005 −0.011 −0.040
(0.001) (0.001) (0.002) (0.020) (0.020) (0.028)
N 2818 2818 2818 2818 2818 2818
Adjusted R2 0.004 0.009 0.010 0.002 0.002 0.002

45
 Table XII
Bond Issuance Cross-Sectional Splits

This table reports the results of the regressions of the difference in bond issuance by matched BB+ and BBB- firms on
high-yield fund flows and differences in the characteristics of matched firms

∆Bond Issuancei,t = α + βF lows · F lowst−1 + X βX · ∆Xi,t−1 + εi,t

�

estimated separately for subsamples of BB+ firms split by firm size, dividends, cash flows, external dependence, having a
bank loan rating, and access to the asset-backed securities market. ∆X = X BB+ − X BBB− is the difference in firm characteristic
X between matched BB+ and BBB- firms. Issuance of non-convertible, not asset- or mortgage-backed bonds from SDC is scaled by
lagged assets. Small firms are the ones below the median of assets for BB+ firms. Dividend payers are firms with positive cash
dividends. Low cash flow firms are the ones below the median of the cash-to-assets ratio for BB+ firms. Appendix Table XX reports
for each Fama-French 48 industry the value of the Rajan and Zingales (1998) measure of external dependence. Industry external
dependence is calculated using data for the 1970-1985 period. Bank loan rating indicates the existence of a bank loan rating.
Top 5 industries by the share of asset- and mortage-backed securities issuance in total industry-level bond issuance are Electrical
Equipment, Personal Services, Automobiles and Trucks, Machinery, and Retail. Appendix Table XXI reports for each Fama-French
48 industry the share of asset- and mortgage-backed issuance in total bond issuance by publicly traded non-financial domestic

46
corporations. The sample period is 1986Q1-2007Q4 in all regressions except for the bank loan rating split, where the sample period
is 1986Q1-2005Q2. Standard errors are adjusted for clustering by both firm and quarter using Thompson (2006). *, **, and ***
denote statistical significance at 10%, 5%, and 1%.
Top 20 External Bank Top 5
Small Size Dividend Payer Low Cash Flow Dependence Industry Loan Rating ABS Industry
Yes No Yes No Yes No Yes No Yes No Yes No
Positive flowst−1 0.009∗∗∗0.007∗ 0.006 0.011∗∗∗ 0.008∗∗ 0.007∗∗ 0.008∗∗ 0.007∗∗ 0.010∗∗∗0.008 0.003 0.009∗∗∗
(0.003) (0.004) (0.004) (0.002) (0.004) (0.003) (0.004) (0.003) (0.003) (0.006) (0.005) (0.003)
∆Qi,t−1 −0.002 −0.001 −0.000 −0.001 0.001 −0.003∗∗ −0.001 −0.001 −0.001 −0.002 0.002 −0.002∗
(0.001) (0.002) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.002) (0.002) (0.001)
∆Log(Assetsi,t−1 ) −0.017∗∗∗−0.009∗ −0.013∗∗∗−0.010∗∗ −0.011∗∗ −0.013∗∗∗−0.012∗∗ −0.012∗∗∗ −0.011∗∗∗ −0.019∗∗−0.016 −0.011∗∗∗
(0.006) (0.005) (0.005) (0.004) (0.005) (0.005) (0.006) (0.004) (0.004) (0.008) (0.010) (0.004)
∆Leveragei,t−1 0.006 −0.002 −0.003 0.004 0.008 −0.010 −0.002 0.005 0.000 0.003 0.018∗ −0.004
(0.012) (0.007) (0.009) (0.008) (0.008) (0.009) (0.009) (0.009) (0.008) (0.014) (0.009) (0.008)
Constant −0.004∗ 0.000 0.001 −0.005∗∗∗−0.001 −0.002 −0.001 −0.002 −0.003 −0.002 −0.001 −0.002
(0.003) (0.001) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.003) (0.003) (0.002)
N 1387 1431 1576 1234 1387 1431 1493 1325 1773 584 583 2235
Adjusted R2 0.015 0.005 0.007 0.015 0.010 0.010 0.009 0.009 0.012 0.008 0.015 0.011
observations without a bank loan rating. Finally, columns 11 and 12 show that firms in non-
ABS industries have a stronger differential sensitivity of issuance to fund flows than firms
in ABS industries. Overall, the small number of issuance events in our sample hampers our
ability to do cross-sectional splits, but the results in Table XII do suggest that financially
constrained firms do have higher sensitivities of issuance to fund flows.
Thus, it appears that bond issuance does play an important role in connecting high yield
mutual fund flows to the investment of BB+ firms. This does not rule out the possibility
that pure cost of capital effects are also at work, but it does help to flesh out the mechanism
driving our results.

V Conclusion

The sharp distinction drawn between investment- and speculative-grade firms is one of
the most salient features of credit markets. These terms are more than convenient labels—we
show that this segmentation has real consequences for firm investment.
Our paper makes four contributions. First, we show that BB+ firms and their BBB-
matches on average have similar investment rates, suggesting that the average allocation
of capital is efficient across segments. Second, we find that flows into high-yield mutual
funds cause the investment of BB+ firms to increase relative to their BBB- matches. Thus,
market segmentation exposes firms to non-fundamental variation in the availability and cost
of capital, which in turn leads to excess volatility in their investment. This is particularly
true for firms that do not have access to other sources of financing. Third, we show that flows
also increase BB+ bond issuance, suggesting that the availability of capital is an important
driver of investment. Finally, we suggest that segmentation facilitates categorization by

47
investors. Fund flows induce excess comovement of the investment of BB+ firms with the
investment of dissimilar, lower-rated firms in unrelated industries.
The distortions induced by fund flows are economically meaningful but not excessively
large. However, our estimates are likely a lower bound because the firms facing the largest
costs of a BB+ rating are likely to alter their behavior to obtain a BBB- rating.
Our work adds to a broader understanding of the structure of capital markets. Our
results should not be interpreted as suggesting that credit ratings are not valuable, or that
the division of the corporate bond market into two grades is not efficient in a broader
sense. Credit ratings carry information and may help investors economize on information
costs. What we want to emphasize is that sharp divides can have significant effects on real
investment that should be weighed carefully against any potential benefits.

48
References
Altman, Edward I., 1968, Financial ratios, discriminant analysis and prediction of corporate
bankruptcy, Journal of Finance 23, 589–609.

, and Duen Li Kao, 1992, The implications of corporate bond ratings drift, Financial
Analysts Journal 48, 64–75.

Baker, Malcolm, 2009, Capital market-driven corporate finance, Annual Review of Financial
Economics 1, 181–205.

, Jeremy C. Stein, and Jeffrey Wurgler, 2003, When does the market matter? Stock
prices and the investment of equity-dependent firms, Quarterly Journal of Economics 118,
969–1005.

Barberis, Nicholas, and Andrei Shleifer, 2003, Style investing, Journal of Financial Eco-
nomics 68, 161–199.

Bernanke, Ben S., and Alan S. Blinder, 1992, The federal funds rate and the channels of
monetary transmission, American Economic Review 82, 901–921.

Bernanke, Ben S., and Mark Gertler, 1995, Inside the black box: The credit channel of
monetary policy transmission, Journal of Economic Perspectives 9, 27–48.

Cantor, Richard, and David T. Hamilton, 2005, Rating transitions and defaults conditional
on rating outlooks revisited: 1995-2005, Discussion paper, Moody’s.

Cantor, Richard, and Chris Mann, 2006, Analyzing the tradeoff between ratings accuracy
and stability, Discussion paper, Moody’s.

Duffie, Darrell, and Bruno Strulovici, 2009, Capital mobility and asset pricing, Working
paper, Stanford University.

Ellul, Andrew, Pab Jotikasthira, and Christian T. Lundblad, 2010, Regulatory pressure and
fire sales in the corporate bond market, Working paper, Indiana University.

Faulkender, Michael, and Mitchell A. Petersen, 2006, Does the source of capital affect capital
structure?, Review of Financial Studies 19, 45–79.

Fazzari, Steven M., R. Glenn Hubbard, and Bruce C. Petersen, 1988, Financing constraints
and corporate investment, Brookings Papers on Economic Activity, pp. 141–195.

Fons, Jerome S., 2004, Tracing the origins of “investment grade”, Special comment, Moody’s.

49
Gertler, Mark, and Simon Gilchrist, 1994, Monetary policy, business cycles, and the behavior
of small manufacturing firms, Quarterly Journal of Economics 109, 309–340.
Gompers, Paul, Joy Ishii, and Andrew Metrick, 2003, Corporate governance and equity
prices, The Quarterly Journal of Economics 118, 107–155.
Harold, Gilbert, 1938, Bond Ratings as an Investment Guide: An Appraisal of Their Effec-
tiveness (The Ronald Press Company: New York).
Hickman, Walter Braddock, 1958, Corporate Bond Quality and Investor Experience (Prince-
ton University Press: Princeton).
Hirano, Keisuke, Guido W. Imbens, and Geert Ridder, 2003, Efficient estimation of average
treatment effects using the estimated propensity score, Econometrica 71, 1161–1189.
Imbens, Guido W., and Thomas Lemieux, 2008, Regression discontinuity designs: A guide
to practice, Journal of Econometrics 142, 615–635.
Imbens, Guido W., and Jeffrey M. Wooldridge, 2009, Recent developments in the economet-
rics of program evaluation, Journal of Economic Literature 47, 5–86.
Jorion, Philippe, Zhu Liu, and Charles Shi, 2005, Informational effects of regulation FD:
Evidence from rating agencies, Journal of Financial Economics 76, 309–330.
Kashyap, Anil K., and Jeremy C. Stein, 2000, What do a million observations on banks say
about the transmission of monetary policy?, American Economic Review 90, 407–428.
, and David W. Wilcox, 1993, Monetary policy and credit conditions: Evidence from
the composition of external finance, American Economic Review 83, 78–98.
Kisgen, Darren J., 2006, Credit ratings and capital structure, Journal of Finance 61, 1035–
1072.
, and Philip E. Strahan, 2009, Do regulations based on credit ratings affect a firm’s
cost of capital, Working paper, Boston College.
Kliger, Doron, and Oded Sarig, 2000, The information value of bond ratings, Journal of
Finance 55, 2879–2902.
Leary, Mark T., 2009, Bank loan supply, lender choice, and corporate capital structure,
Journal of Finance 64, 1143–1185.
Lemmon, Michael L., and Michael R. Roberts, 2009, The response of corporate financing
and investment to changes in the supply of credit, Journal of Financial and Quantitative
Analysis forthcoming.

50
Levich, Richard M., Giovanni Majnoni, and Carmen M. Reinhart (eds.), 2002, Ratings,
Rating Agencies and the Global Financial System (Kluwer Academic Publishers: Boston).

Mayer, Colin, 1988, New issues in corporate finance, European Economic Review 32, 1167–
1183.

Mitchell, Mark, Lasse Heje Pedersen, and Todd Pulvino, 2007, Slow moving capital, Amer-
ican Economic Review 97, 215–220.

Moody’s, 2009, Corporate default and recovery rates, 1920-2008, Special comment.

Mullainathan, Sendhil, 2000, Thinking through categories, Working paper, MIT.

Polk, Christopher, and Paola Sapienza, 2009, The stock market and corporate investment:
A test of catering theory, Review of Financial Studies 22, 187–217.

Rajan, Raghuram G., and Luigi Zingales, 1998, Financial dependence and growth, American
Economic Review 88, 559–586.

Standard & Poor’s, 2008, Corporate ratings criteria 2008, Technical report.

Sufi, Amir, 2009, Bank lines of credit in corporate finance, Review of Financial Studies 22,
1057–1088.

Tang, Tony T., 2009, Information asymmetry and firms’ credit market access: Evidence from
Moody’s credit rating format refinement, Journal of Financial Economics 93, 325–351.

Thompson, Samuel B., 2006, Simple formulas for standard errors that cluster by both firm
and time, Working paper, Harvard University, Department of Economics.

U.S. Securities and Exchange Commission, 2003, Report on the role and function of credit
rating agencies in the operation of the securities markets, Special report required by Section
702(b) of the Sarbanes-Oxley Act of 2002.

West, Richard R., 1973, Bond ratings, bond yields, and financial regulation: Some findings,
Journal of Law and Economics 16, 159–168.

51
VI Appendix

Table XIII
Variable Definitions

Variable Definition
Book leverage Book debt divided by the sum of book debt and stockholder equity.
Cash flow Income before extraordinary items plus depreciation. Cash flow is annualized.
Coverage Interest coverage is the ratio of EBIT to interest expense, calculated using four-
quarter moving averages of EBIT and interest expense.
Credit spread Difference in yields between Moody’s Baa and Aaa rated industrial bonds. Aver-
age of the end-of-month values (from the Federal Reserve Statistical Release H.15
“Selected Interest Rates”) during quarter t.
Flowst−1 Monthly aggregate flows into high-yield mutual funds are from the Investment
Company Institute. Cumulative flows over the four quarters [t − 4, t − 1] are
scaled by the total PPE of firms rated BBB+ through BB-, P P Et−1 . The value
of fund flows is standardized so that the coefficient on flows represents the effect
of a one standard deviation change in high-yield fund flows.
GDP growth Percentage change during quarter t in the seasonally adjusted real GDP (from the
Bureau of Economic Analysis).
CAP Xi,t
Investment Capital expenditures scaled by lagged PPE, P P Ei,t−1 . Investment is annualized.
Market leverage Book debt divided by the sum of book debt and market value of equity from
CRSP.
Operating margin Operating income before depreciation divided by sales.
Q Market value of equity from CRSP plus assets minus the book value of stockholder
equity, all divided by assets.
ROA Income before extraordinary items divided by assets. ROA is annualized.
Stock market return Value-weighted return on all NYSE, AMEX, and NASDAQ stocks minus the 1-
month T-Bill rate. Average of monthly values (from Kenneth French’s website)
during quarter t.
Term spread Difference in yields between 10-year constant-maturity Treasuries and 3-month
T-Bills. Average of the end-of-month values (from the Federal Reserve Statistical
Release H.15 “Selected Interest Rates”) during quarter t.
VIX Chicago Board Options Exchange Volatility Index. Average of the end-of-month
values (from Datastream) during quarter t.
Z-score 1.2 · W Ci,t /Assetsi,t + 1.4 · REi,t /Assetsi,t + 3.3 · EBITi,t /Assetsi,t +
Salesi,t /Assetsi,t , where W Ci,t is working capital, and REi,t is retained earn-
ings.

52
 8,000
Number of firm-quarter observations

6,000

4,000

2,000

0
AAA AA A+ A- BBB BB+ BB- B CCC+ CCC- CC
AA+ AA- A BBB+ BBB- BB B+ B- CCC CC+ C

Figure 2
Distribution of Issuer Credit Ratings

This figure shows the distribution of S&P domestic long-term issuer credit rating for firms in the
quarterly CRSP/Compustat merged data set, excluding financials and utilities. The sample period is
1986Q1-2007Q4.

53
 Table XIV
Default Rates by Credit Rating

Panel A reports 1-, 5-, and 10-year default rates by credit rating. Each column of Panel B reports the default rate for
firms with the specified credit rating relative to the default rate for firms rated one notch higher. For example, column BB+ reports
the ratio of BB+ and BBB- default rates. CCC includes firms rated CCC+ through C. Data from the Standard & Poor’s 2007
Annual Global Corporate Default Study and Rating Transitions. The sample period is 1981-2007.
Investment Grade Speculative Grade

54
AAA AA+ AA AA- A+ A A- BBB+ BBB BBB- BB+ BB BB- B+ B B- CCC
Panel A: Default Rates
1 year 0.00 0.00 0.00 0.02 0.05 0.07 0.06 0.15 0.23 0.31 0.52 0.81 1.44 2.53 6.27 9.06 25.59
5 years 0.28 0.20 0.18 0.45 0.61 0.60 0.73 1.74 1.95 3.74 5.41 8.38 12.32 17.65 23.84 29.44 44.50
10 years 0.67 0.35 0.72 1.08 1.46 1.73 2.12 3.71 4.44 6.91 10.21 14.62 21.03 26.11 30.43 35.73 49.76
Panel B: Ratios of Default Rates
1 year 2.50 1.40 0.86 2.50 1.53 1.35 1.68 1.56 1.78 1.76 2.48 1.44 2.82
5 years 0.71 0.90 2.50 1.36 0.98 1.22 2.38 1.12 1.92 1.45 1.55 1.47 1.43 1.35 1.23 1.51
10 years 0.52 2.06 1.50 1.35 1.18 1.23 1.75 1.20 1.56 1.48 1.43 1.44 1.24 1.17 1.17 1.39
 Table XV
Differences in Firm Characteristics across Credit Ratings

Each column reports the difference between firms with the specified credit rating and firms rated one notch higher. For ex-
ample, column BB+ reports the difference between firms rated BB+ and firms rated BBB-. Panel A reports differences scaled by the
X̄1 −X̄0
absolute value of the mean of firms rated one notch higher. Panel B reports normalized differences ∆X = √ 2 2
, where Sg2 is the
S0 +S1
sample variance of X for group g. Firms rated AAA and AA+ are treated as one credit rating. Firms rated CCC+ through CCC-
are also treated as one credit rating. The sample of firms in the quarterly CRSP/Compustat merged data set, excluding financials
and utilities. The sample period is 1986Q1-2007Q4.
Investment Grade Speculative Grade
AA AA- A+ A A- BBB+ BBB BBB- BB+ BB BB- B+ B B- CCC
Panel A: Scaled Difference
Assets −0.288 −0.329 −0.047 −0.139 −0.159 −0.083 −0.215 −0.123 −0.278 −0.377 −0.262 −0.287 0.496 −0.236 −0.152
Book leverage 0.192 0.054 0.095 0.035 0.007 0.070 0.066 0.044 0.083 0.093 0.078 0.111 0.138 −0.008 0.191
Market leverage 0.147 0.115 0.154 0.154 0.165 0.059 0.218 0.038 0.116 0.107 0.119 0.133 0.123 −0.009 0.215
Z-score −0.077 −0.008 −0.006 −0.079 −0.093 −0.094 −0.044 −0.045 −0.097 −0.064 −0.158 −0.251 −0.836 −3.344 −1.769
Interest coverage −0.344 −0.094 −0.111 −0.160 −0.200 −0.037 −0.260 0.085 −0.201 −0.136 −0.161 −0.454 −0.603 −1.788 −0.198

55
Cash/Assets −0.115 −0.165 −0.077 0.132 −0.118 −0.026 −0.079 0.185 0.026 0.071 0.151 0.170 0.338 0.302 −0.350
PPE/Assets 0.045 −0.107 0.004 −0.028 0.098 0.045 −0.051 −0.098 0.090 −0.056 −0.018 −0.026 0.016 −0.021 0.177
Q −0.027 −0.087 −0.057 −0.066 −0.083 −0.015 −0.132 0.036 −0.039 −0.014 −0.014 −0.042 0.031 0.071 −0.090
Operating margin −0.136 −0.132 0.040 −0.055 0.039 0.047 −0.184 0.037 0.024 −0.054 0.084 −0.191 −0.325 −0.889 2.084
ROA −0.093 −0.153 −0.005 −0.162 −0.129 −0.059 −0.194 −0.038 −0.105 −0.156 −0.383 −1.077 NM −1.165 −0.343
CF/PPE −0.117 −0.091 0.021 −0.023 −0.003 −0.088 −0.065 0.331 −0.108 0.069 −0.168 −0.189 −0.686 −2.921 −0.422
Capex/PPE −0.029 0.004 −0.011 −0.004 0.023 0.029 −0.103 0.109 −0.009 0.103 0.006 0.052 0.082 0.020 −0.381
Panel B: Normalized Difference
Assets −0.222 −0.226 −0.024 −0.072 −0.078 −0.036 −0.101 −0.061 −0.150 −0.213 −0.136 −0.125 0.095 −0.065 −0.051
Book leverage 0.213 0.071 0.129 0.054 0.011 0.110 0.107 0.062 0.109 0.124 0.109 0.160 0.190 −0.011 0.257
Market leverage 0.106 0.096 0.140 0.159 0.175 0.068 0.241 0.045 0.136 0.131 0.151 0.179 0.175 −0.013 0.317
Z-score −0.167 −0.017 −0.011 −0.157 −0.163 −0.147 −0.062 −0.056 −0.105 −0.063 −0.140 −0.171 −0.370 −0.211 −0.237
Interest coverage −0.295 −0.065 −0.075 −0.101 −0.120 −0.020 −0.151 0.039 −0.096 −0.058 −0.063 −0.173 −0.167 −0.249 −0.033
Cash/Assets −0.084 −0.110 −0.051 0.071 −0.067 −0.013 −0.043 0.093 0.014 0.037 0.074 0.086 0.168 0.166 −0.268
PPE/Assets 0.060 −0.169 0.006 −0.040 0.115 0.053 −0.063 −0.113 0.091 −0.062 −0.018 −0.027 0.016 −0.023 0.176
Q −0.040 −0.134 −0.095 −0.106 −0.120 −0.021 −0.220 0.052 −0.058 −0.022 −0.021 −0.064 0.041 0.080 −0.108
Operating margin −0.230 −0.205 0.051 −0.070 0.042 0.044 −0.185 0.031 0.020 −0.044 0.060 −0.138 −0.158 −0.223 0.052
ROA −0.109 −0.159 −0.005 −0.152 −0.094 −0.036 −0.115 −0.017 −0.041 −0.050 −0.096 −0.150 −0.240 −0.239 −0.134
CF/PPE −0.099 −0.075 0.020 −0.019 −0.002 −0.053 −0.038 0.138 −0.052 0.028 −0.072 −0.067 −0.175 −0.219 −0.059
Capex/PPE −0.043 0.006 −0.015 −0.004 0.025 0.031 −0.110 0.095 −0.008 0.079 0.005 0.038 0.056 0.013 −0.287
 Table XVI
Inertia in Credit Ratings

This table documents inertia in credit ratings by comparing the characteristics of BB+ firms before they are upgraded with
the characteristics of all BBB- firms and with the characteristics of BBB- before they are downgraded and by similarly comparing
the characteristics of BBB- firms before they are downgraded with the characteristics of all BB+ firms. Panel A compares the
characteristics of BB+ firms in two years before they are upgraded with the characteristics of all BBB- firms. Panel B compares the
characteristics of BBB- firms in two years before they are downgraded with the characteristics of all BB+ firms. Panel C compares
the characteristics of BB+ firms in two years before they are upgraded with the characteristics of BBB- firms in two years before
they are downgraded. Cash, PPE, cash flow, and capital expenditures are scaled by assets. The sample period is 1986Q1-2007Q4.
Standard errors are adjusted for clustering by firm. *, **, and *** denote statistical significance at 10%, 5%, and 1%.

Leverage
Assets Book Market Z-score Coverage Cash PPE Q Margin ROA CF Capex
Panel A: BB+ (N = 850) before upgrade versus all BBB- (N = 5377)
BBi,t + −1009 0.022 −0.033∗∗ −0.033 −0.388 0.012 0.001 0.184∗∗∗ 0.005 0.006∗∗∗ 0.026∗∗∗ 0.000
(628) (0.023) (0.014) (0.052) (0.741) (0.008) (0.022) (0.063) (0.011) (0.001) (0.005) (0.005)
Constant 5677∗∗∗ 0.456∗∗∗ 0.313∗∗∗ 0.824∗∗∗ 6.881∗∗∗ 0.064∗∗∗ 0.347∗∗∗ 1.566∗∗∗ 0.164∗∗∗ 0.010∗∗∗ 0.089∗∗∗ 0.063∗∗∗

56
(506) (0.016) (0.009) (0.033) (0.501) (0.004) (0.015) (0.039) (0.008) (0.001) (0.003) (0.003)
N 6227 5985 5985 5520 6005 6227 6216 6227 6218 6227 6227 6227
Adjusted R2 0.001 0.001 0.004 0.000 −0.000 0.002 −0.000 0.006 0.000 0.012 0.012 −0.000
Panel B: BBB- (N = 1155) before downgrade versus all BB+ (N = 3894)
BBi,t + −2932∗∗ 0.006 −0.042∗∗∗ −0.007 0.379 0.014∗∗∗ 0.030∗ 0.162∗∗∗ 0.038∗∗∗ 0.005∗∗∗ 0.028∗∗∗ 0.011∗∗∗
(1202) (0.017) (0.015) (0.042) (0.844) (0.005) (0.017) (0.050) (0.010) (0.001) (0.004) (0.004)
Constant 6988∗∗∗ 0.486∗∗∗ 0.388∗∗∗ 0.749∗∗∗ 5.200∗∗∗ 0.052∗∗∗ 0.348∗∗∗ 1.351∗∗∗ 0.129∗∗∗ 0.004∗∗∗ 0.062∗∗∗ 0.060∗∗∗
(1404) (0.016) (0.014) (0.043) (0.799) (0.004) (0.019) (0.051) (0.009) (0.001) (0.004) (0.004)
N 5049 4866 4866 4467 4883 5046 5041 5049 5041 5049 5049 5049
Adjusted R2 0.007 −0.000 0.008 −0.000 0.000 0.005 0.003 0.009 0.012 0.010 0.019 0.004
Panel C: BB+ (N = 850) before upgrade versus BBB- before downgrade (N = 1155)
BBi,t + −2320 −0.009 −0.108∗∗∗ 0.042 1.294 0.025∗∗∗ 0.001 0.398∗∗∗ 0.040∗∗∗ 0.012∗∗∗ 0.053∗∗∗ 0.003
(1428) (0.024) (0.016) (0.053) (1.012) (0.007) (0.023) (0.072) (0.011) (0.001) (0.005) (0.005)
Constant 6988∗∗∗ 0.486∗∗∗ 0.388∗∗∗ 0.749∗∗∗ 5.200∗∗∗ 0.052∗∗∗ 0.348∗∗∗ 1.351∗∗∗ 0.129∗∗∗ 0.004∗∗∗ 0.062∗∗∗ 0.060∗∗∗
(1405) (0.016) (0.014) (0.043) (0.799) (0.004) (0.019) (0.051) (0.009) (0.001) (0.004) (0.004)
N 2005 1925 1925 1735 1949 2005 2001 2005 2000 2005 2005 2005
Adjusted R2 0.003 −0.000 0.081 0.001 0.004 0.024 −0.000 0.061 0.023 0.102 0.107 0.000
 Table XVII
Robustness to Alternative Scaling of Investment

This table shows the robustness of our results to alternative scaling of investment. Capital expendi-
tures are alternatively scaled by assets, assets net of cash, or PPE. Cash flow is scaled using the same
variable as investment. All variables are standardized so that the coefficients represent the effect in standard
deviations of the dependent variable of one standard deviation change in an explanatory variable. Fund
flows are calculated over quarters [t − 4, t − 1] and scaled by fund total net assets, T N At−5 . Interactions
of firm characteristics with fund flows are included in columns 2, 4, and 6, but are not reported. The
sample period is 1986Q1-2007Q4. Standard errors are adjusted for clustering by both firm and quarter
using Thompson (2006). *, **, and *** denote statistical significance at 10%, 5%, and 1%.

Assets Assets - Cash PPE

(1) (2) (3) (4) (5) (6)
∆Qi,t−1 0.036 0.038 0.039 0.043 0.092∗∗ 0.079∗
(0.035) (0.058) (0.040) (0.063) (0.037) (0.047)
∆CFi,t 0.258∗∗∗ 0.254∗∗∗ 0.248∗∗∗ 0.246∗∗∗ 0.295∗∗∗ 0.238∗∗∗
(0.039) (0.047) (0.043) (0.051) (0.055) (0.049)
Flowsi,t−1 0.048∗ 0.056∗ 0.050∗∗ 0.060∗∗ 0.057∗∗ 0.069∗∗∗
(0.026) (0.030) (0.025) (0.029) (0.022) (0.025)
∆Log(Assetsi,t−1 ) −0.060∗ −0.047 −0.059∗ −0.047 −0.063∗∗ −0.065∗∗
(0.034) (0.035) (0.035) (0.035) (0.031) (0.032)
∆Leveragei,t−1 −0.040 −0.080 −0.057 −0.085∗ −0.120∗∗∗ −0.140∗∗∗
(0.037) (0.051) (0.038) (0.050) (0.033) (0.040)
∆Z-scorei,t−1 −0.091∗∗ −0.079∗ 0.026
(0.041) (0.042) (0.034)
∆Coveragei,t−1 0.033 0.039 0.057∗
(0.033) (0.034) (0.034)
Constant −0.000 0.019 −0.000 0.019 0.000 0.022
(0.044) (0.046) (0.044) (0.047) (0.038) (0.040)
N 2818 2147 2817 2147 2818 2147
Adjusted R2 0.094 0.113 0.095 0.113 0.162 0.163

57
 Table XVIII
Differences in the Characteristics of Firms in Placebo Matched Samples

This table reports the differences in the characteristics of firms in placebo matched samples in Table VIII. In each column
firms with the specified credit rating that are below the 75th percentile of market leverage for that credit rating are matched with
firms rated one notch higher. The sample period is 1986Q1-2007Q4.Standard errors are adjusted for clustering by firm. *, **, and
*** denote statistical significance at 10%, 5%, and 1%.

A A- BBB+ BBB BBB- BB+ BB BB- B+ B

Assets -68 364 145 49 268 275 -19 7 43 101
Book leverage 0.012 0.015 0.009 0.027 0.028

58
−0.017 −0.005 −0.019 −0.031 −0.002
Market leverage −0.011 −0.002 0.011 0.003 −0.041∗∗ −0.006 −0.016 −0.013 0.001 −0.018
Z-score −0.061 −0.122∗∗ −0.054 −0.053 −0.001 −0.098∗ −0.070 −0.113∗∗ −0.131∗∗∗ −0.287∗∗∗
Interest coverage 0.507 −1.569 −3.318∗ −1.660∗ 1.188 −0.875 −1.184 −0.731 −1.398∗∗∗ −1.043∗∗
Cash/Assets 0.019∗∗ −0.002 −0.006 −0.007 0.014∗∗ 0.001 0.027∗∗∗ 0.014∗ 0.008 0.055∗∗∗
PPE/Assets −0.005 0.012 0.010 −0.028 −0.021 0.012 −0.013 0.001 −0.005 −0.013
Q 0.126 −0.048 −0.306∗ −0.163∗∗ 0.078 0.014 0.053 −0.032 −0.013 0.131∗∗
Operating margin −0.011 0.004 −0.013 −0.036∗∗∗ 0.001 0.001 −0.012 0.007 −1.008 −5.918
ROA −0.000 −0.001 −0.001 −0.001 0.001 0.000 −0.002 −0.003∗∗ −0.004∗∗∗ −0.006∗∗
CF/PPE 0.017 0.078 −0.075 −0.024 0.126∗∗∗ 0.010 −0.028 −0.122 −0.048 −0.142∗∗
Capex/PPE 0.009 0.003 0.001 −0.002 0.025∗∗ 0.006 0.035∗∗ −0.001 0.022∗ 0.043∗∗
 Table XIX
Propensity Score Weighting

This table reports the results of the regressions of firm investment on fund flows interacted with
speculative-grade rating. Junki,t is a binary variable equal to one for firms rated BB+. In columns 3-5,
BB+ observations are weighted by p(X) 1
and BBB- observations are weighted by 1−p(X) 1
. Propensity score,
p(X), is estimated as a logistic regression of Junki,t on firm characteristics. Column 1 reports the results
of the propensity score estimation. Column 2 reports the results of unweighted regressions. Columns 3-5
report the results of propensity score weighted regressions. Industry-quarter fixed effects are included in
the investment regressions in columns 2-5. In column 5 interactions of firm characteristics with fund flows
are included but not reported. The sample period is 1986Q1-2007Q4. Cumulative high-yield mutual fund
flows over the four quarters [t − 4, t − 1] are scaled by the total PPE of all firms rated BBB+ through BB-,
P P Et−1 . The value of fund flows is standardized so that the coefficient on flows represents the effect of one
standard deviation change in fund flows. Investment and cash flow are annualized. In column 1, standard
errors are adjusted for clustering by firm. In columns 2-5, standard errors are adjusted for clustering by
both firm and quarter using Thompson (2006). *, **, and *** denote statistical significance at 10%, 5%,
and 1%.

Propensity Score Unweighted Propensity Score Weighted

(1) (2) (3) (4) (5)
Qi,t−1 −0.026 0.042∗∗∗ 0.039∗∗∗ 0.030∗∗∗ 0.031∗∗∗
(0.111) (0.013) (0.013) (0.010) (0.011)
CFi,t
P P Ei,t−1 0.044∗∗∗ 0.048∗∗∗ 0.043∗∗∗ 0.043∗∗∗
(0.010) (0.010) (0.010) (0.011)
Junki,t 0.009 0.009 0.011 0.011
(0.010) (0.010) (0.010) (0.010)
Junki,t ·Flowst−1 0.017∗∗ 0.014∗ 0.013∗ 0.013∗
(0.008) (0.008) (0.007) (0.007)
Log(Assetsi,t−1 ) −0.326∗∗∗ −0.004 −0.004
(0.064) (0.005) (0.005)
Book leveragei,t−1 0.209 −0.065∗ −0.067∗
(0.471) (0.034) (0.034)
Market leveragei,t−1 1.068∗ −0.100∗∗ −0.098∗∗
(0.638) (0.044) (0.044)
N 8380 8721 8380 8380 8380
Adjusted R2 0.030 0.289 0.279 0.303 0.303

59
 Table XX
External Dependence by Fama-French 48 Industry

This table reports the value of the Rajan and Zingales (1998) measure of external dependence by
Fama-French 48 industry. Using annual Compustat data set covering the 1970-1985 period, we first
calculate for each firm total capital expenditures minus total cash flows from operations during this period,
all scaled by total capital expenditures. We then take the industry median as the industry measure of
external dependence.

External External
Industry Dependence Industry Dependence
1 Agriculture 0.09 23 Automobiles and Trucks −0.39
2 Food Products −0.40 24 Aircraft −0.41
3 Candy & Soda −0.28 25 Shipbuilding −0.38
4 Beer & Liquour −0.34 26 Defense −0.51
5 Tobacco Products −1.18 27 Precious Metals 0.63
6 Recreation −0.28 28 Mining −0.17
7 Entertainment 0.29 29 Coal −0.11
8 Printing and Publishing −0.90 30 Petroleum and Natural Gas 0.56
9 Consumer Goods −0.44 32 Communication 0.06
10 Apparel −1.02 33 Personal Services 0.15
11 Healthcare 0.27 34 Business Services −0.11
12 Medical Equipment 1.04 35 Computers 0.26
13 Pharmaceutical Products 0.74 36 Electronic Equipment 0.01
14 Chemicals −0.42 37 Measuring and Control Equipment −0.31
15 Rubber and Plastic Products −0.24 38 Business Supplies −0.29
16 Textiles −0.33 39 Shipping Containers −0.24
17 Construction Materials −0.31 40 Transportation 0.23
18 Construction −0.24 41 Wholesale −0.26
19 Steel Works −0.14 42 Retail −0.14
20 Fabricated Products −0.20 43 Restaurants, Hotels, Motels 0.35
21 Machinery −0.43 48 Other 0.32
22 Electrical Equipment −0.43

60
 Table XXI
Asset-Backed and Mortgage-Backed Bond Issuance by Fama-French 48 Industry

This table reports the share of asset- and mortage-backed bonds in total non-convertible bond is-
suance by Fama-French 48 industry. Non-convertible bond issuance by domestic publicly-traded firms is
from SDC. Shelf registrations and initiations of medium-term note programs are excluded. The sample
period is 1986Q1-2007Q4.

ABS ABS
Industry share (%) Industry share (%)
1 Agriculture 0.00 23 Automobiles and Trucks 19.46
2 Food Products 0.22 24 Aircraft 5.85
3 Candy & Soda 0.32 25 Shipbuilding 0.00
4 Beer & Liquour 0.00 26 Defense 0.09
5 Tobacco Products 0.00 27 Precious Metals 0.00
6 Recreation 4.37 28 Mining 0.00
7 Entertainment 0.24 29 Coal 0.00
8 Printing and Publishing 0.48 30 Petroleum and Natural Gas 2.65
9 Consumer Goods 7.33 32 Communication 0.68
10 Apparel 0.12 33 Personal Services 26.86
11 Healthcare 0.51 34 Business Services 5.08
12 Medical Equipment 0.00 35 Computers 0.67
13 Pharmaceutical Products 0.10 36 Electronic Equipment 0.25
14 Chemicals 0.31 37 Measuring and Control Equipment 0.00
15 Rubber and Plastic Products 0.43 38 Business Supplies 0.12
16 Textiles 0.77 39 Shipping Containers 2.06
17 Construction Materials 7.55 40 Transportation 3.16
18 Construction 10.22 41 Wholesale 1.19
19 Steel Works 2.06 42 Retail 16.37
20 Fabricated Products 4.19 43 Restaurants, Hotels, Motels 2.49
21 Machinery 17.78 48 Other 18.82
22 Electrical Equipment 31.89

61