This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Research and Development, Uncertainty, andAnalysts’ Forecasts: The Case of IAS 38
Tami DinhInstitute of Accounting, Control and Auditing, University of St. Gallen, Tigerbergstr. 9,9000, St. Gallen, Switzerlande-mail: [email protected]
Brigitte EierleDepartment of International Accounting and Auditing, University of Bamberg,Feldkirchenstr. 21, 96045, Bamberg, Germanye-mail: [email protected]
Wolfgang SchultzeDepartment of Accounting and Control, University of Augsburg, Universitaetsstr. 16,86159, Augsburg, Germanye-mail: [email protected]
Leif SteegerDepartment of Accounting and Control, University of Augsburg, Universitaetsstr. 16,86159, Augsburg, Germanye-mail: [email protected]
Abstract
This study analyzes the consequences of the capitalization of development expendituresunder IAS 38 on analysts’ earnings forecasts. We use unique hand-collected data in asample of highly research and development (R&D)-intensive German-listed firms overthe period 2000–2007. We find that the capitalization of development costs is signifi-cantly associated with both higher individual analysts’ forecast errors and forecast dis-persion. This suggests that the increasing complexity surrounding the capitalization ofdevelopment costs negatively impacts forecast accuracy. However, for firms with highunderlying environmental uncertainty, forecast errors are negatively associated withcapitalized development expenditures. This indicates that the negative impact ofincreased complexity on forecast accuracy can be outweighed by the information con-tained in the signals from capitalized development costs when the underlying environ-mental uncertainty is high. The findings contribute to the ongoing controversial debate
We gratefully acknowledge helpful comments by Thomas Guenther, Alan Hodgson, BaruchLev, Thorsten Sellhorn, Brian Singleton-Green, Marco Wilkens, Anne Wyatt, Stefano Zambon,and workshop participants at the University of Augsburg, the annual EAA congress in Istanbul,and the EIASM workshop on Visualising, Measuring and Managing Intangibles & IntellectualCapital in Ferrara.
Data availability: Data are available from public sources identified in this study.
Journal of International Financial Management & Accounting 26:3 2015
on the accounting for self-generated intangible assets. Our results provide usefulinsights on the link between capitalization of development costs, environmental uncer-tainty, and analysts’ forecasts for accounting academics and practitioners alike.
1. Introduction
This study analyzes the consequences of capitalizing development
expenditures under IAS 38 for analysts’ forecast accuracy and forecast
dispersion. The standard requires capitalizing development costs under
specific, restrictive conditions.1 As a consequence, not all research and
development (R&D) is capitalized but only an a priori unknown frac-
tion. Analysts need to forecast not only future amounts of R&D, but
also capitalization rates, amortization rates on previously capitalized
amounts, and potential write-offs. Consequently, the forecasting com-
plexity increases with the capitalization of development expenditures.
Analysts’ forecast accuracy has been found to decrease with increasing
forecast complexity (see Ramnath et al., 2008 for a review). Accord-
ingly, Aboody and Lev (1998) find that the complexities involved in
the capitalization process lead to higher analysts’ forecast errors for
firms who capitalize parts of their software development costs. This is
consistent with other studies showing that R&D is highly uncertain
and complex, making it hard to obtain accurate analysts’ forecasts
(Amir et al., 2003; Chambers et al., 2003; Gu and Wang, 2005).
In other settings, however, the evidence points to the contrary.
Some studies find evidence for higher forecast accuracy (Anagnosto-
poulou, 2010) associated with capitalized development costs. Similarly,
research has found higher value relevance associated with R&D capi-
talization (Lev and Sougiannis, 1996) and lower information asymme-
tries for capitalized software development (Mohd, 2005). Matolcsy and
Wyatt (2006) find that capitalization of intangible assets is associated
with lower earnings forecast dispersion and lower absolute earnings
forecast error. These benefits from capitalization derive from the dis-
cretion involved in the capitalization decision, which allows managers
to signal their private information on the prospects of the investment
to the market (Matolcsy and Wyatt, 2006; Ahmed and Falk, 2009).
The evidence in Wyatt (2005) suggests that management is capitalizing
when their firm has more certain intangible investments, that is, invest-
ments with less uncertain future benefits and, therefore, more predict-
able earnings (Matolcsy and Wyatt, 2006).
258 Tami Dinh, Brigitte Eierle, Wolfgang Schultze and Leif Steeger
This table displays Tobit regression results on the determinants of capitalizing developmentcosts (equation 3) using Huber/White adjusted standard errors for Tobit regression (z-statis-tics in parentheses). All variables are defined as outlined in the appendix.Two-tailed significance: *p < .10, **p < .05, ***p < .01.
Research and Development, Uncertainty, and Analysts’ Forecasts 279
This table shows empirical results related to hypotheses 1a and 1b. The results from Table 2are used here to estimate the instrumented value of DCAP. This table presents the 2SLSresults for forecast accuracy (first column) and forecast dispersion (second column) separatelyusing Huber/White sandwich estimators to estimate robust standard errors. Pooled sample Ais used in the regression with individual analysts’ forecast errors BFE as the dependent vari-able. Panel sample A0 is used in the regression with forecast dispersion SDF as the dependentvariable, calculated on the basis of individual analysts’ forecast errors.Two-tailed significance: *p < .10, **p < .05, ***p < .01.
280 Tami Dinh, Brigitte Eierle, Wolfgang Schultze and Leif Steeger
and forecast dispersion. In addition, we show that the negative impact
of increased complexity can be outweighed by the information con-
tained in the signals from capitalization of development costs when the
underlying environmental uncertainty is high. The interpretation of the
information contained in capitalized development costs seems to vary
across different levels of underlying economic uncertainty. Our findings
contribute to the discussion of the role of analysts as financial interme-
diaries in the capital market and may help in understanding their use
of information and process of dissemination more closely (Ramnath
et al., 2008).
We acknowledge a number of caveats in our study. The German
setting allows us to hold country-specific changes constant. However,
we are aware that the findings might differ in environments where ana-
lysts are historically more familiar with interpreting the information
contained in capitalized development expenditures (e.g., United King-
dom, France, and Australia). A cross-cultural analysis might provide
useful insights into that matter.
We do not—apart from the count variable of years of IFRS—explicitly account for behavioral aspects concerning how analysts pro-
cess information, for example, a possible learning effect from a change
in accounting rules or herding behavior. We encourage future research
to consider such behavioral aspects and to analyze our research ques-
tion using an experimental research design. This allows drawing fur-
ther conclusions on how capitalization of development costs under
IAS 38 affects the forecasting process of market participants, in partic-
ular of analysts.
Notes
1. IAS 38 requires firms to capitalize development costs from the point in time whenthe criteria in IAS 38.57 are cumulatively met. Research costs as well as developmentcosts incurred before the criteria are met are expensed.2. Even though IAS 38 requires firms to capitalize development costs, the applicationof the criteria in IAS 38.57 involves significant discretion (e.g., Leibfried and Pfanzelt,2004; Meyer and Naumann, 2009).3. Our observation period ends in 2007 to avoid confounding with the recent financialcrisis (e.g., Kang et al., 2014). Our expectations are based on the conjecture that firmsreact to higher levels of business risk by a greater use of discretionary accruals and thatat the same time, the resulting signals are more informative. Hence, we are interested inthe effects of economic risk affecting management’s choice to recognize intangibleassets. As the financial crisis was an exogenous shock to the entire financial system andcaused large distortions in the global economy, it is not the type of uncertainty we areinvestigating. Rather, the resulting distortions conceal the underlying economic risks of
Research and Development, Uncertainty, and Analysts’ Forecasts 287
a firm’s regular operations and would likely bias our analyses. It is unlikely that R&Dcapitalizations would resolve the uncertainties caused by the financial crisis. To the con-trary, Francis et al. (2013) find conservative accounting practices, such as expensingR&D, to be an important governance mechanism ensuing less declines in stock pricesduring the financial crisis. Therefore, we exclude the time period of the financial crisisfrom our analysis.4. This is consistent with Hope (2004) who reveals that forecast accuracy also deterio-rates if the forecasting process is complicated by extensive accounting choices.5. However, recent work by Ciftci (2010) also suggests that the capitalization of soft-ware development under SFAS 86 (ASC 350–40) reduces earnings quality.6. As SIZE and FOL have been found to be of concern for multicollinearity (Botosan,1997), we only use one control variable of the two at a time. In our sample, SIZE andFOL show a correlation of 75 per cent with firm size measured by the natural loga-rithm of market value of equity. The reported results include FOL, but our resultsremain unaffected when using SIZE instead (negative regression coefficient with p-value < .01). Similarly to FOL, SIZE also proxies for disclosure quality (Bailey et al.,2003).7. Our results remain qualitatively unchanged when running the regressions for eachquintile of volatility instead of sextile.8. In our analyses on forecast dispersion, the sample size is fairly small due to the cal-culation of standard deviation of forecast errors. Hence, we focus the analysis usingsextiles of uncertainty on forecast errors only (H1c).9. To avoid survivorship bias, all acquired or failed companies during the observationperiod remain in the sample even when data are not available for subsequent periods.10. We also run our main regression models without firms belonging to the GeneralStandard and the results remain unchanged.11. The negative sign for IFRS implies that companies that just recently moved toIFRS are more likely to capitalize more development costs than companies that havebeen using IFRS for some time already. Under German GAAP (HGB), the capitaliza-tion of development expenditures was strictly prohibited. Anecdotal evidence suggeststhat the possibility to capitalize development costs was one of the main drivers for anumber of German companies to voluntarily adopt IFRS pre-2005.12. Note that the last column in Panel A of Table 4 shows 255 observations at thepanel level (instead of 248) because for some firm years, we have information on thedecision to capitalize (dummy_CAP) but not on the amount of capitalized developmentexpenditures (DCAP).13. Note that BIG is significant and positive in the pooled sample, which is not consis-tent with our expectations. This may be due to the large dominance of firms in oursample that are audited by a big auditor (about 90 per cent).
References
Aboody, D. and B. Lev, “The Value Relevance of Intangibles: The Case of SoftwareCapitalization,” Journal of Accounting Research 36 (Suppl.) (1998), pp. 161–191.
Aboody, D. and B. Lev, “Information Asymmetry, R&D, and Insider Gains,” TheJournal of Finance 55 (2000), pp. 2747–2766.
Ahmed, K. and H. Falk, “The Riskiness of Future Benefits: The Case of Capitalizationof R&D and Capital Expenditures,” Journal of International Accounting Research 35(2009), pp. 45–60.
288 Tami Dinh, Brigitte Eierle, Wolfgang Schultze and Leif Steeger
Akerlof, G.A., “The Market for “Lemons”: Quality Uncertainty and the Market Mech-anism,” The Quarterly Journal of Economics 84 (1970), pp. 488–500.
Alford, A. and P. Berger, “A Simultaneous Equations Analysis of Forecast Accuracy,Analyst Following, and Trading Volume,” Journal of Accounting, Auditing andFinance 14 (1999), 219–240.
Amir, E., Y. Guan and G. Livne, “The Association of R&D and Capital Expenditureswith Subsequent Earnings Variability,” Journal of Business Finance and Accounting34 (2007), pp. 222–246.
Amir, E., B. Lev and T. Sougiannis, “Do Financial Analysts Get Intangibles?” Euro-pean Accounting Review 12 (2003), pp. 635–659.
Anagnostopoulou, S.C., “Does the Capitalization of Development Costs Improve Ana-lyst Forecast Accuracy? Evidence from the UK,” Journal of International FinancialManagement and Accounting 21 (2010), pp. 62–83.
Ashbaugh, H. and M. Pincus, “Domestic Accounting Standards, InternationalAccounting Standards, and the Predictability of Earnings,” Journal of AccountingResearch 39 (2001), pp. 417–434.
Bailey, W., H. Li, C.X. Mao and R. Zhong, “Regulation Fair Disclosure and EarningsInformation: Market, Analyst, and Corporate Responses,” Journal of Finance 58(2003), pp. 2487–2514.
Barron, O., D. Byard, C. Kile and E. Riedl, “High-Technology Intangibles and Ana-lysts’ Forecasts,” Journal of Accounting Research 40 (2002), pp. 289–312.
Barth, M., R. Kasznik and M. McNichols, “Analyst Coverage and Intangible Assets,”Journal of Accounting Research 39 (2001), pp. 1–34.
Behn, B.K., J.-H. Choi and T. Kang, “Audit Quality and Properties of Analyst Earn-ings Forecasts,” The Accounting Review 83 (2008), pp. 327–349.
Botosan, C.A., “Disclosure Level and the Cost of Equity Capital,” The AccountingReview 72 (1997), pp. 323–349.
Boujelbene, M.A. and H. Affes, “The Impact of Intellectual Capital Disclosure on Costof Equity Capital: A Case of French Firms,” Journal of Economics, Finance andAdministrative Science 18 (2013), pp. 45–53.
Cazavan-Jeny, A. and T. Jeanjean, “The Negative Impact of R&D Capitalization: AValue Relevance Approach,” European Accounting Review 15 (2006), pp. 7–61.
Cazavan-Jeny, A., T. Jeanjean and P. Joos, “Accounting Choice and Future Perfor-mance: The Case of R&D Accounting in France,” Journal of Accounting and PublicPolicy 30 (2011), pp. 145–165.
Chambers, D., R. Jennings and R.B. Thompson, “Managerial Discretion and Account-ing for Research and Development Costs,” Journal of Accounting, Auditing andFinance 18 (2003), pp. 79–113.
Cheng, J.L.C. and I.F. Kesner, “Organizational Slack and Response to EnvironmentalShifts: The Impact of Resource Allocation Patterns,” Journal of Management 23(1997), pp. 68–86.
Ciftci, M., “Accounting Choice and Earnings Quality: The Case of Software Develop-ment,” The European Accounting Review 19 (2010), pp. 429–459.
Danielson, M.G. and E. Press, “When Does R&D Expense Distort Profitability Esti-mates?” Journal of Applied Finance 15 (2005), pp. 76–92.
Dichev, I.D. and V.W. Tang, “Earnings Volatility and Earnings Predictability,” Journalof Accounting and Economics 47 (2009), pp. 160–181.
Dinh, T. and W. Schultze, “Economic Consequences of Discretionary, Partial R&DCapitalization under IAS 38,” Conference Paper EAA Istanbul (2010).
Duru, A. and D. Reeb, “International Diversification and Analysts’ Forecast Accuracyand Bias,” The Accounting Review 77 (2002), pp. 415–433.
Research and Development, Uncertainty, and Analysts’ Forecasts 289
Dye, R., “An evaluation of “Essays on Disclosure” and the Disclosure Literature inAccounting,” Journal of Accounting & Economics 32 (2001), pp. 181–235.
Fisher, I., The Theory of Interest, 1st edn (New York: The MacMillan Company,1930).
Francis, B., I. Hasan and Q. Wu, “The Benefits of Conservative Accounting to Share-holders: Evidence from the Financial Crisis,” Accounting Horizons 27 (2013), pp.319–346.
Ghosh, D. and L. Olsen, “Environmental Uncertainty and Managers’ Use of Discre-tionary Accruals,” Accounting, Organizations and Society 34 (2009), pp. 188–205.
Glaum, M., J. Baetge, A. Grothe and T. Oberdoerster, “Introduction of InternationalAccounting Standards, Disclosure Quality and Accuracy of Analysts’ Earnings Fore-casts,” European Accounting Review 22 (2013), pp. 79–116.
Govindarajan, V.J., “Appropriateness of Accounting Data in Performance Evaluation:An Empirical Examination of Environmental Uncertainty as an Intervening Vari-able,” Accounting, Organizations and Society 13 (1984), pp. 125–136.
Graham, J.R., C. Harvey and S. Rajgopal, “The Economic Implications of CorporateFinancial Reporting,” Journal of Accounting and Economics 40 (2005), pp. 3–73.
Gu, F. and W. Wang, “Intangible Assets, Information Complexity and Analysts’ Fore-casts Earnings,” Journal of Business, Finance and Accounting 32 (2005), pp. 1673–1702.
Hirst, E., P.E. Hopkins and J.M. Wahlen, “Fair Values, Income Measurement, andBank Analysts’ Risk and Valuation Judgments,” The Accounting Review 79 (2004),pp. 454–473.
Hodgdon, C., R. Tondkar, D. Harless and A. Adhikari, “Compliance with IFRS Dis-closure Requirements and Individual Analysts’ Forecast Errors,” Journal of Interna-tional Accounting, Auditing and Taxation 17 (2008), pp. 1–13.
Hope, O.-K., “Variations in the Financial Reporting Environment and Earnings Fore-casting,” Journal of International Financial Management and Accounting 15 (2004),pp. 22–43.
Hwang, L., C. Jan and S. Basu, “Loss Firms and Analysts’ Earnings Forecast Errors,”Journal of Financial Statement Analysis 1 (1996), pp. 18–30.
Kang, T., I.K. Khurana and C. Wang, “International Diversification, SFAS 131 andPost-Earnings Announcement Drift,” (2014), Working paper Oklahoma State Uni-versity, University of Missouri, Florida International University.
Kennedy, P., A Guide to Econometrics, 6th edn (Ann Arbor, MI: South – Western Col-lege Publishers, 2008).
Kothari, S.P., T. Laguerre and A. Leone, “Capitalization Versus Expensing: Evidenceon the Uncertainty of Future Earnings from Capital Expenditures Versus R&D Out-lays,” Review of Accounting Studies 7 (2002), pp. 355–382.
Leibfried, P. and S. Pfanzelt, “Praxis der Bilanzierung von Forschungs- und Ent-wicklungskosten gem€aß IAS/IFRS – Eine empirische Untersuchung deutscher Unter-nehmen [Accounting of Research and Development Expenditures According to IAS/IFRS in Practice – An Empirical Analysis of German Companies],” Zeitschrift f€ur in-ternationale und kapitalmarktorientierte Rechnungslegung (KoR) 4 (2004), pp. 491–497.
Lambert, R.A., C. Leuz and R.E. Verrecchia, “Accounting Information, Disclosure,and the Cost of Capital,” Journal of Accounting Research 45 (2007), pp. 385–420.
Lev, B. and T. Sougiannis, “The Capitalization, Amortization, and Value Relevance ofR&D,” Journal of Accounting and Economics 21 (1996), pp. 107–138.
290 Tami Dinh, Brigitte Eierle, Wolfgang Schultze and Leif Steeger
Louis, H. and D. Robinson, “Do Managers Credibly Use Accruals to Signal PrivateInformation? Evidence from the Pricing of Discretionary Accruals Around StockSplits,” Journal of Accounting and Economics 39 (2005), pp. 361–380.
Markarian, G., L. Pozza and A. Prencipe, “Capitalization of R&D Costs and EarningsManagement: Evidence from Italian Listed Companies,” The International Journal ofAccounting 43 (2008), pp. 246–267.
Matolcsy, Z. and A. Wyatt, “Capitalized Intangibles and Financial Analysts,” Account-ing & Finance 46 (2006), pp. 457–479.
Meyer, H. and K.P. Naumann, “Enforcement der Rechnungslegung – aktuelle Ent-wicklungen [Enforcement of Accounting – Current Developments],” Die Wir-tschaftspr€ufung 62 (2009), pp. 807–811.
Mohd, E., “Accounting for Software Development Costs and Information Asymme-try,” The Accounting Review 80 (2005), pp. 1211–1231.
OECD, “Main Science and Technology Indicators Database,” (2009).Orens, R., W. Aerts and N. Lybaert, “Intellectual Capital Disclosure, Cost of Financeand Firm Value,” Management Decision 47 (2009), pp. 1536–1554.
Oswald, D., “The Determinants and Value Relevance of the Choice of Accounting forResearch and Development Expenditures in the United Kingdom,” Journal of Busi-ness, Finance and Accounting 35 (2008), pp. 1–24.
Oswald, D. and P. Zarowin, “Capitalization of R&D and the Informativeness of StockPrices,” European Accounting Review 16 (2007), pp. 703–726.
Peek, E., “The Influence of Accounting Changes on Financial Analysts’ Forecast Accu-racy and Forecasting Superiority: Evidence from the Netherlands,” EuropeanAccounting Review 14 (2005), pp. 261–295.
Plumlee, M., “The Effect of Information Complexity on Analysts’ Use of That Infor-mation,” The Accounting Review 78 (2003), 275–296.
Pownall, G. and K. Schipper, “Implications of Accounting Research for the SEC’sConsideration of International Accounting Standards for U.S. Securities Offerings,”Accounting Horizons 13 (1999), pp. 259–280.
Ramnath, S., S. Rock and P. Shane, “The Financial Analyst Forecasting Literature: ATaxonomy with Suggestions for Future Research,” International Journal of Forecast-ing 24 (2008), pp. 34–75.
Riley, J.G., “Competitive Signalling,” Journal of Economic Theory 10 (1975), 174–186.Riley, J.G., “Silver Signals: Twenty-Five Years of Screening and Signalling,” Journal ofEconomic Literature 39 (2001), pp. 432–478.
Skinner, D., “Accounting for Intangibles – a Critical Review of Policy Recommenda-tions,” Accounting and Business Research 38 (2008), 191–204.
Subramanyam, K.R., “The Pricing of Discretionary Accruals,” Journal of Accountingand Economics 22 (1996), pp. 249–281.
Tan, H.C. and B. Sidhu, “Sources of Earnings Variability and Their Effect on EarningsForecasts,” Accounting & Finance 52 (1 Suppl.) (2012), pp. 343–371.
Teoh, S.H. and C.Y. Hwang, “Nondisclosure and Adverse Disclosure as Signals ofFirm Value,” The Review of Financial Studies 4 (1991), pp. 283–313.
Tucker, J.W. and P. Zarowin, “Does Income Smoothing Improve Earnings Informa-tiveness?” The Accounting Review 81 (2006), pp. 251–270.
Research and Development, Uncertainty, and Analysts’ Forecasts 291
Umanath, N., M. Ray and T. Campbell, “The Effect of Uncertainty and InformationAsymmetry on the Structure of Compensation Contracts: A Test of Competing Mod-els,” Management Science 42 (1996), pp. 868–874.
Verrecchia, R.E., “Discretionary Disclosure,” Journal of Accounting and Economics 5(1983), pp. 179–194.
Verrecchia, R.E., “Essays on Disclosure,” Journal of Accounting and Economics 32(2001), pp. 97–180.
Webster, E., The Economics of Intangible Investment, 1st edn (Cheltenham: Edward El-gar Publishing, 1999).
Weiss, D., H. Falk and U.B. Zion, “Earnings Variability and Disclosure of R&D: Evi-dence from Press Releases,” Accounting & Finance 53 (2013), pp. 837–865.
Wrigley, J., “Discussion of ‘What Financial and Non-Financial Information on Intan-gibles is Value-Relevant? A Review of the Evidence,” Accounting and BusinessResearch 38 (2008), pp. 257–260.
Wyatt, A., “Accounting and Recognition of Intangible Assets: Theory and Evidence onEconomic Determinants,” The Accounting Review 80 (2005), pp. 967–1003.
Wyatt, A., “What Financial and Non-Financial Information on Intangibles is ValueRelevant? A Review of the Evidence,” Accounting and Business Research 38 (2008),pp. 217–256.
Appendix A
Dependent variables
BFEit = natural logarithm of absolute individual analyst forecast error
(= difference between actual earnings per share and forecasted earnings
per share by a financial analyst, scaled by end of year share price).
SDFit = natural logarithm of standard deviation of analysts’ forecast,
scaled by end of year share price.
Independent variables related to R&D
dummy_CAPit = binary variable equal to 1 if a firm capitalizes devel-
opment expenditures on the balance sheet, 0 otherwise.
DCAPit = annual capitalized development expenditures deflated by
market value at the end of a year.
DCAPlagit = annual capitalized development expenses in period t-1
deflated by market value of equity at the end of year.
RDGit = absolute change in total R&D expenditures relative to the
prior period as a growth measure.
RDINTit = R&D intensity for firm i in year t computed as R&D