Skip to main content
SpringerLink
Log in

Does CDS market price intangible asset value? Evidence from SG&A expenditure

  • Original Research
  • Published:
Review of Quantitative Finance and Accounting Aims and scope Submit manuscript

Abstract

Prior studies document that SG&A expenditure (exclusive of R&D and advertising expenditure) creates an intangible asset value despite that U.S. GAAP requires SG&A to be expensed immediately as a period cost. In this paper, we investigate whether the credit default swap (CDS) market recognizes the intangible asset value of SG&A expenditure. Our findings show that although the CDS market views contemporaneous SG&A expenditure as a risk factor on average, it does recognize the intangible asset value created by SG&A. Specifically, the positive association between CDS spread and SG&A expenditure is weaker for firms with higher SG&A intangible asset value. This suggests that SG&A’s intangible asset value helps mitigate credit risk. We also document that such mitigation effect is stronger when the accounting information of the reference entity is more transparent and when the reference entity is more financially constrained. Our inferences remain when we: (1) examine the short window CDS market reaction to SG&A information around SEC filing dates, and (2) use total SG&A expenditure including R&D and advertising in our analysis. Overall, this study shows that the credit market differentiates between the expense and the asset components of SG&A expenditure and lowers the risk assessment of firms with higher intangible asset values.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Firms report SG&A expenditure in income statement as the sum of all direct expenses and indirect expenses not related to the production of any goods or service. Specifically, selling expenditure includes the cost of marketing, distributing products and providing product-related services. General and administrative expenditure includes the costs of managing and developing the business (Banker et al. 2019). Banker et al. (2019) show that the SG&A expenditure on average accounts for 37 percent of a company’s total assets, compared to 4 percent for R&D expenditure in their sample.

  2. To clarify, SG&A expenditure and R&D/ advertising expense are very different in at least the following two respects. Firstly, SG&A expenditures are more commonly reported than R&D/advertising across all industries, and the magnitude is usually greater (Banker et al. 2019). More importantly, the intangibles commingled in SG&A significantly exceed R&D and advertising (Enache and Srivastava 2018). Prior research on the value of intangible assets has mainly focused on research and development (R&D) and advertising expenditures (Woolridge 1988; Chan, Martin, and Kensinger 1990; Lev and Sougiannis 1996; Chan, Lakonishok, and Sougiannis 2001; Eberhart, Maxwell, and Siddique 2004), and they provide scant details on its constituent items other than R&D and advertising (Enache and Srivastava 2018). Given that SG&A expenditure creates future benefits, the accounting treatment for SG&A, especially the portion excluding R&D and advertising, may lead to investors’ under-valuation of SG&A (Banker et al. 2019). Secondly, the future benefits created by SG&A are more certain (i.e., less risky) than that of R&D (Enache and Srivastava 2018). Compared to equity market investors, debt market investors are more risk averse and care more about the downside risk (Myers 1977, Acharya et al. 2011). Therefore, given different characteristics of SG&A and R&D/advertising expense, CDS market may price them differently. We would like to thank our referee for bring this to our attention.

  3. For instance, marketing activities such as brand developing, customer relationship maintaining, and management group training create future value in different ways, depending on firms’ nature (Leone and Schultz 1980; Cleland and Bruno 1996; Hauser et al. 1994; Ittner and Larcker 1998; Banker et al. 2000; Brynjolfsson and Hitt 2000).

  4. Recent studies also document that excess stock returns can be earned by holding portfolios of companies with high SG&A intangible asset value (Anderson et al. 2007; Eisfeldt and Papanikolaou 2013; Banker et al. 2019).

  5. Unlike standard structural models (Merton 1974) and reduced-form models (Das 1995; Das and Sundaram 2000; Hull and White 2000a, 2000b), the DL model relaxes the assumption that the asset value and asset structure of reference entities are all observable to credit market investors when they explore the role of financial reports in the credit market pricing process. In their model, investors receive noisy accounting reports about firms’ asset dynamics and asset values, reflecting imperfect information about the bankruptcy risk of reference entities.

  6. In the DL model, the price of a CDS contract is a function of the firm’s assets and leverage. However, the asset value and asset dynamics reflected by periodic financial statements may contain noise. Therefore, the CDS market will also consider other information that is related to firm’s overall bankruptcy risk in the pricing process. For example, Callen et al. (2009) find that the CDS market offers a discount on CDS contracting price when earnings are favorable and demands a higher CDS premium when earnings are unfavorable. The CDS market may have a better estimate on the intrinsic asset value of the company when the company has a better information environment.

  7. The earliest paper discussing CDS contracting price is Merton (1974). In that paper, the author proposes a structural model and relates CDS price to firm’s asset value and capital structure. However, strong assumptions are imposed by the structural model. First, the assets of the reference entity are completely observable. Second, the asset structure of the reference entity behaves according to a known stochastic process. These two assumptions are not likely to be satisfied in the real world. The reduced form model proposed at late 1990s tries to fix the caveat of the structural model, but it does not do a good job. Specifically, the reduced form model releases the full asset revelation assumption of the structural model, but it imposes another assumption that the uncertainty on firm’s asset valuation only comes from exogenous factors. The assumption of the reduced form model counters the fact that the uncertainly on assets could be endogenously determined to a large extent in reality. Therefore, the DL model is considered as more advanced in explaining the reality compared to previous two models, and widely used as a basic theoretical framework in academic research.

  8. Quite a few follow-up empirical studies start to test the implications of DL model and show evidences on the role of accounting signals not reflected in the balance sheet in CDS pricing. For example, a stream of studies examines income statement items and document a negative association between CDS premium and earnings (Callen et al. 2009). Their explanation is that higher earnings indicate higher profitability and thereby lower probability of default in the future (Benkert 2004, Batta 2006, Callen et al. 2009, Das et al. 2009).

  9. A twenty percent straight-line amortization rate is commonly used to estimate the contribution of R&D to firm value every year, so we include four lags of R&D expenditure (deflated by total assets) (Griliches 1979; Chan et al. 2001; Hall et al. 1988). Prior literature has documented that the benefits entailed by advertising investments are usually reflected in firm value soon through its impact on sales in the near future, so we include one lag of advertising expenditure in the estimation model (deflated by total assets) (Pale 1970; Banker et al. 2011; 2019).

  10. The simultaneous problem arises in our estimation model when a firm-level idiosyncratic shock affects both dependent and independent variables and causes inconsistent estimates of coefficients. For instance, an upgrade on technology may affect a firm’s current earnings, current SG&A spending and the effectiveness of SG&A investment simultaneously. To address the simultaneous problem, we use the average SG&A expenditure (deflated by total assets) of other firms in firm i’s four digit SIC code industry as an instrument of the firm’s SG&A expenditure. Industry-average SG&A expenditure is highly correlated with a firm’s SG&A spending but uncorrelated with firm-level idiosyncratic shocks. This practice follows prior studies (Lev and Sougiannis 1996; Hanlon et al. 2003, Banker et al. 2019), as industry-average SG&A expenditure is highly correlated with a firm’s SG&A spending but uncorrelated with firm-level idiosyncratic shocks. To further clarify, we only use industry average as instrument for SG&A when we estimate SG&A future value (model (1), to construct variable SG&A Future Value), but do not use the instrument for the SG&A in our main test (\(SG\&A/T{A}_{t}\) of model (2)).

  11. We require at least 10 non-missing observations to obtain the firm-specific estimate of SG&A future value.

  12. We eliminate observations with at least one non-missing value on SG&A, R&D and Advertising expenditures in the preceding four years.

  13. We obtain qualitatively similar results when using 8 percent or 12 percent as the discount rate.

  14. We need firms’ industry information to construct the instrumental variable, which is the average SG&A (deflated by total assets) of other firms in the same four-digit SIC code industry.

  15. Because we require non-missing data on all variables (including lagged variables) and at least ten time-series observations in the rolling window to perform the time-series regression, we can only get estimates of SG&A future value from 1995 which is thirteen years after 1982.

  16. For a subsample of firm-quarters with non-negative SG&A Future Value, the mean (median) value of SG&A Future Value is 1.150 (0.169). Our main results are robust to the subsample analysis of non-negative SG&A Future Value.

  17. Untabulated results suggests that our inference stays the same when we partition the sample using different cutoffs of Z-scores.

  18. In untabulated analyses, we further estimate the future value of R&D and advertising expense with the same method and explore their risk mitigation effects respectively. The results show that neither R&D nor advertising future value has a significant risk mitigation effect on CDS market’s pricing, which is consistent with our expectation given the stark differences between SG&A vs. R&D and advertising expense. The estimation strategy of the paper may not fully capture the underlying value of R&D and advertising expense given their high uncertainty. We thank our referee for the suggestion.

References

Download references

Acknowledgements

We acknowledge support from National Natural Science Foundation of China (72172038, 72241425), Sci-Tech Innovation Foundation of School of Management, Fudan University, and Research Fund of Shanghai Data Exchange Corporation.

Author information

Authors and Affiliations

  1. Fudan University, Shanghai, China

    Rong Huang

  2. Central University of Finance and Economics, Beijing, China

    Xintian Lin

  3. Fordham University, Bronx, USA

    Yuan Xie

Authors
  1. Rong Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xintian Lin
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yuan Xie
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Xintian Lin.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix

Variable definitions

 

Variable definitions

TA

Total assets (Compustat quarterly item ATQ)

SG&A/TA

Selling, general and administrative expenditure exclusive of R&D and advertising expenditures (Compustat quarterly items XSGAQ—XRDQ—XAD), deflated by total assets

Total SG&A/TA

Total selling, general and administrative expenditure (Compustat quarterly item XSGAQ), deflated by total assets

R&D/TA

R&D expenditure (Compustat quarterly item XRDQ), deflated by total assets

ADV/TA

Advertising expenditure deflated by total assets. Since the quarterly data of advertising expenditure is not available in Compustat, advertising expenditure equals to the annual advertising expense (Compustat annual item XAD) for the last quarter of each fiscal year and equals to 0 for the first three quarters

OI /TA

Operating income before depreciation, SG&A, R&D, and advertising expenditures (Compustat quarterly items OIBDPQ + XSGAQ), deflated by total assets

CDS Spread

The spread of 5-year credit default swap, measured in basis points, from Markit

DEL_CDSSpread

The percentage change of the CDS spread, measured in percentage points

DEPTH

Number of contributors providing quotes in a CDS trading day from Markit

MV

Market value (Compustat quarterly items PRCCQ*CSHOQ)

ROA

Income before extraordinary items, SG&A, R&D, and advertising expenditures (Compustat quarterly items IBQ + XSGAQ), deflated by total assets

LEV

The sum of short-term and long-term debt (Compustat quarterly items DLTTQ + TLCQ) divided by the sum of total liability (COMPUSTAT quarterly item LTQ) and market value of assets (MV)

SD_RET

The standard deviation of company’s daily equity returns over the fiscal quarter

SPOT

The annualized 3-month Treasury-Bill rate from Compustat

RATING

S&P domestic long-term issuer credit rating from Compustat. The higher is the Rating, the lower the credit rating quality

Institutional Ownership

The percentage of institutional holdings reported during the quarter. The variable is constructed based on Thomson-Reuters 13F (TR-13F) S34 TYPE3 holding data

Altman Z-Score

Altman Z-Score is constructed following Altman (1968):

\(Altman\,\mathrm{ Z}-\mathrm{Score}=1.2 \times \left(\frac{Net\, working\, capital \left(ACTQ-LCTQ\right)}{Total assets \left(ATQ\right)}\right) +1.4 \times \left(\frac{Retained\, earnings (REQ)}{Total\, assets (ATQ)}\right)+3.3 \times \left(\frac{Earnings\, before\, interest\, and\, taxes(PIQ+XINTQ)}{Total\, assets (ATQ)}\right)+0.6 \times \left(\frac{Market\, value\, of\, equity (MV)}{Book\, value\, of\, liabilities (LTQ)}\right)+1.0\times (\frac{Sales (SALEQ)}{Total\, assets (ATQ)})\)

Variables measuring value-creating ability of SG&A expenditure

SG&A Future Value

The future value-creating ability of SG&A expenditure, which is the sum of discounted coefficients on past SG&A (\(\sum_{k=1}^{3}\frac{{\alpha }_{2,k}}{{\left(1.1\right)}^{k}}\)), estimated from the following model for each firm-year using a rolling window of time-series data (annual data) starting from 1982

\(\begin{aligned} \left( {\frac{{OI}}{{TA}}} \right)_{{i.t}} = & \alpha _{0} + \alpha _{1} \left( {\frac{1}{{TA}}} \right)_{{i,t - k}} + \sum\limits_{{K = 0}}^{3} {\alpha _{{2,k}} } \left( {\frac{{SG\& A}}{{TA}}} \right)_{{i,t - k}} + \sum\limits_{{K = 0}}^{4} {\alpha _{{3,k}} } \left( {\frac{{R\& D}}{{TA}}} \right)_{{i,t - k}} \\ & \quad + \alpha _{4} \left( {\frac{{ADV}}{{TA}}} \right)_{{i,t}} + \varepsilon _{{i,t}} \\ \end{aligned}\)

Details of the estimation procedure are provided in Sect. 3

Total SG&A Future Value

The future value-creating ability of Total SG&A expenditure, which is the sum of discounted coefficients on past SG&A (\(\sum_{k=1}^{3}\frac{{\alpha }_{2,k}}{{\left(1.1\right)}^{k}}\)) estimated from the following model for each firm-year using a rolling window of time-series data (annual data) starting from 1982

\({\left(\frac{OI}{TA}\right)}_{i.t}= {\alpha }_{0}+{\alpha }_{1}{\left(\frac{1}{TA}\right)}_{i,t-k}+\sum_{K=0}^{3}{\alpha }_{2,k}{\left(\frac{Total SG\&A}{TA}\right)}_{i,t-k} +{\varepsilon }_{i,t}\)

Details of the estimation procedure are provided in Sect. 3

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Huang, R., Lin, X. & Xie, Y. Does CDS market price intangible asset value? Evidence from SG&A expenditure. Rev Quant Finan Acc 61, 701–728 (2023). https://doi.org/10.1007/s11156-023-01165-0

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11156-023-01165-0

Keywords

Search

Navigation