Does the Riskiness of R&D Outweigh Its Benefits? The Perspective of US Private Lenders

• Author: Mustafa Ciftci,Masako Darrough
• DOI: http://doi.org/10.1111/jbfa.12196
• Published date: 01 May 2016
• Date: 01 May 2016

Journal of Business Finance & Accounting

Journal of Business Finance & Accounting, 43(5) & (6), 654–692, May/June 2016, 0306-686X

doi: 10.1111/jbfa.12196

Does the Riskiness of R&D Outweigh Its

Benefits? The Perspective of US Private

Lenders

MUSTAFA CIFTCI AND MASAKO DARROUGH∗

Abstract: We investigate the relationship between R&D investments and loan spread. Prior

research documents that R&D is associated with greater future benefits and risks, suggesting

that the valuation of R&D depends on a tradeoff between the two. Some research finds that

bondholders consider that the benefits of R&D outweigh its risks: R&D is negatively associated

with bond yields. This is surprising given that debt holders are more concerned about downside

risk due to asymmetric payoffs. Using data on private debt from the US, we find an overall

positive association between loan spread and R&D intensity, suggesting that the riskiness of

R&D appears to outweigh its benefits for private lenders. Furthermore, an asymmetric payoff

structure implies that the risks of R&D for lenders increase with default risk. Consistent with

this argument, we find a positive association between R&D and loan spread for firms that are

smaller, with high default-risk ratings, unrated (no public debt), or in industries with weaker

legal protection. Unrated firms are in the most R&D-intensive group and make up nearly 60%

of the firms with private debt. Consequently, studies that exclude unrated firms are likely to

present an incomplete picture of the perspective of debt holders on R&D.

Keywords: future benefits and risks of R&D, loan pricing, private debt, default risk

1. INTRODUCTION

Prior research suggests that the economy is shifting from tangibles to intangibles (e.g.,

intellectual property, goodwill, brand recognition). Thus, intangible investments such

as research and development (R&D) activities are playing a more important role

in today’s economy than capital expenditures. Lev (2001) suggests that abnormal

profits and dominant competitive positions are achieved by the sound deployment

of intangible assets, enabling firms to earn monopoly rents over a period of time.

Prior research suggests that R&D-intensive firms generate abnormal earnings levels

and growth (Chan et al., 2001; and Eberhart et al., 2004). However,there is also a large

degree of technical and commercial uncertainty associated with R&D investments. The

success rate of innovation projects is often very low, and the distribution of payoffs is

highly skewed (Lev, 2001). For example, in the pharmaceutical industry in the US,

∗The first author is from the American University of Sharjah, UAE. The second author is from Baruch

College-CUNY, USA. (Paper received May 2015, revised revision accepted February 2016).

Address for correspondence: Mustafa Ciftci, School of Business and Management, American University of

Sharjah, PO Box 26666, Sharjah, UAE.

e-mail: mciftci@aus.edu

C

2016 John Wiley & Sons Ltd 654

DOES THE RISKINESS OF R&D OUTWEIGH ITS BENEFITS? 655

only five in 5,000 compounds that enter preclinical testing make it to human testing,

and only one of these five tested compounds is approved by the Food and Drug

Administration (Wierenga and Eaton, 1993). While the US pharmaceutical companies

spend over $50 billion on R&D per year, the number of new drugs that are approved

annually has not increased: in 2008, 21 new drugs were approved for marketing in the

US (Munos, 2009). Consistent with the greater uncertainty associated with R&D, prior

research documents that R&D is associated with a greater variability in future earnings

(Kothari et al., 2002) and returns (Chan et al., 2001).

Valuation of R&D by equity investors and debt holders depends on the tradeoff

between future benefits and risks of R&D. Prior research documents that R&D is

associated with positive contemporaneous returns (Lev and Sougiannis, 1996), which

suggests that equity investors perceive that the future benefits of R&D outweigh its

risks. Although uncertainty associated with R&D activities affects both equity investors

and debt holders, its effects are not the same. Debt holders benefit less from unlimited

upside potential than equity investors because their returns are restricted to the fixed

interest rate and the face value of the debt that the borrower has agreed to pay.

However, they bear the full extent of the downside risk just as equity investors do.

Consequently, debt holders are more concerned about downside risk, whereas equity

investors are more concerned about upside potential (Easton et al., 2009). That is,

the tradeoff between the future benefits and risks of R&D for debt holders is different

from that for equity investors.

Prior research on debt holders’ perspective on R&D has examined public debt

to ask whether higher levels of R&D benefit bondholders. Shi (2003) is the first

study to examine the impact of R&D intensity (defined as R&D expense/market

capitalization) on bond risk premiums and ratings. He finds a negative (positive)

relationship between R&D intensity and bond ratings (bond premium). By contrast,

Eberhart et al. (2008) argue that these relationships change their signs when R&D

intensity is measured with sales or total assets as the deflator: that is, premiums on

bonds decrease with R&D intensity. Hence, prior research focuses only on public debt

and finds conflicting results. However, it does not provide a complete picture of the

debt holders’ perspective on R&D investment because it ignores private debt.

R&D firms that can issue public debt are usually larger firms; many smaller firms

are precluded from public debt financing. Thus, studies that use only public debt

exclude many small, start-up R&D firms that might better represent the population of

innovative firms. As a result, these studies are not able to fully examine the risk-return

tradeoff that smaller R&D-intensive firms face. Of course, a large cross-section of R&D

firms accesses capital via bank loans. Large and small, young and mature, and high and

low R&D-intensity firms in various industries all use bank loans either exclusively or in

addition to public debt. Conclusions based on public debt may not hold true when

considering the perspective of private lenders. It is important, therefore, to include

all types of firms in any investigation. For example, studying only public debt could

severely under-represent smaller firms with potentially different benefit/risk profiles;

the risks and future benefits of R&D activities are subject to scale effects (Ciftci and

Cready, 2011). In this paper, we shift our focus from public debt to private debt and

explore the relationship between R&D activity and loan spread to better understand

the private lender’s perspective on R&D activities.

We use US data from Compustat and DealScan for the period from 1988 to 2011. As

part of the lending process, lenders assess the expected future benefits and riskiness of

C

2016 John Wiley & Sons Ltd

656 CIFTCI AND DARROUGH

R&D projects. This tradeoff between benefit and risk will manifest itself in loan spreads

in the form of a risk premium. We expect the risk premium to increase, on average,

with R&D intensity, controlling for various firm and loan characteristics. We define

R&D intensity as the ratio of R&D capital (capitalized R&D expenditures over time,

net of amortization) to the sum of total assets and R&D capital. We use R&D capital

rather than the current-period R&D expense to focus on R&D as invested capital.

To control for profitability, we use pre-R&D return on assets (PRERND ROA)asthe

primary measure.1We define PRERND ROA as income before extraordinary items plus

R&D expense also deflated by the sum of total assets and R&D capital. We also use

different definitions of R&D intensity and profitability in one of the sensitivity tests.

Contrary to Eberhart et al. (2008), we find that the association between R&D

intensity and loan spread is positive and significant. On average, loan spread increases

with R&D intensity. To illustrate the economic significance of the increase, we estimate

that lenders charge higher spreads (23 bps, on average) for firms in the most R&D-

intensive quintile compared to non-R&D firms. To test the robustness of this finding

to our choice of profitability measure, we use alternative definitions of profitability

(rather than PRERND ROA); we still find a positive association between R&D intensity

and loan spread in all specifications except one. The only exception is ROA, which

produces an insignificant association between R&D intensity and loan spread. This

insignificant result is likely to be the result of multicollinearity, caused by the high

negative correlation between ROA and R&D intensity. Overall, our results suggest that

there is a positive association between loan spread and R&D. Hence, we conjecture that

the negative association between bond yields and R&D documented in prior literature

(Eberhart et al., 2008) could be driven by the selection bias arising from using only

public debt; public debt firms are larger and better established.

The positive association between R&D and loan spread is limited to internally gen-

erated R&D activities, however. Neither goodwill nor separately reported purchased

intangibles are associated with loan spread. This is probably because (1) goodwill is the

result of M&A activities (e.g., overpayment), and may have little to do with intangibles

that generate future profitability; and (2) purchased intangibles are less likely to carry

any residual technical uncertainty.

While perceived risk increases, on average, with R&D intensity, the positive associ-

ation we find masks the possible cross-sectional variation in the relationship. In order

to explore cross-sectional variation, we partition our sample into subgroups based

on default risk, size, and the degree to which R&D outputs are protected from cut-

throat competition. We first investigate how default risk affects the association between

R&D and loan spread. Lenders face greater uncertainty about the repayment of their

loans by high-default-risk firms (Easton et al., 2009) than by low-default-risk firms.

Therefore, we predict that the association between R&D and loan spread will be

greater for firms with a higher default risk. We measure a firm’s default risk with its

bond rating (if rated, otherwise ‘unrated’). We find that R&D is positively associated

with loan spread only for ‘high’ default-risk (speculative-grade) and unrated firms, but

is not significantly associated with loan spread for ‘low’ default-risk (investment-grade)

firms. Since unrated firms make up close to 60% of our sample and tend to be more

R&D intensive and smaller than rated firms, it is essential that these firms be included

when we study the relationship between debt and R&D. Studies that focus solely on

1 Since we are treating R&D expenditures as investment, we measure profitability before R&D expenses.

C

2016 John Wiley & Sons Ltd