Journal of accounting, auditing finance tập 26, số 01, 2011 1

Journal of Accounting, Auditing & Finance http://jaf.sagepub.com/ The Importance of Accounting Information in Portfolio Optimization John R M Hand and Jeremiah Green Journal of Accounting, Auditing & Finance 2011 26: DOI: 10.1177/0148558X11400577 The online version of this article can be found at: http://jaf.sagepub.com/content/26/1/1 Published by: http://www.sagepublications.com On behalf of: Sponsored by The Vincent C Ross Institute of Accounting Research, The Leonard N Stern School of Business Additional services and information for Journal of Accounting, Auditing & Finance can be found at: Email Alerts: http://jaf.sagepub.com/cgi/alerts Subscriptions: http://jaf.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://jaf.sagepub.com/content/26/1/1.refs.html >> Version of Record - Jan 1, 2011 What is This? Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 The Importance of Accounting Information in Portfolio Optimization JOHN R M HAND* JEREMIAH GREEN* We study the economic importance of accounting information as defined by the value that a sophisticated investor can extract from publicly available financial statements when optimizing a portfolio of U.S equities Our approach applies the elegant new parametric portfolio policy method of Brandt, Santa-Clara, and Valkanov (2009) to three simple and firmspecific annual accounting characteristics—accruals, change in earnings, and asset growth We find that the set of optimal portfolio weights generated by accounting characteristics yield an out-of-sample, pre-transactionscosts annual information ratio of 1.9 as compared to 1.5 for the standard price-based characteristics of firm size, book-to-market, and momentum We also find that the delevered hedge portion of the accounting-based optimal portfolio was especially valuable during the severe bear market of 2008 because unlike many hedge funds it delivered a hedged return in 2008 of 12 percent versus only percent for price-based strategies and À38 percent for the value-weighted market Keywords: Accounting Trading Strategies, Portfolio Optimization Introduction The majority of accounting research gauges the importance of accounting information to equity investors by either its usefulness in fundamental analysis and firm valuation (Penman [2009]), or the degree to which a publicly observed accounting signal is able to predict future abnormal stock returns (Bernard & Thomas [1989, 1990]; Lee [2001]) In this paper, we instead define the economic importance of accounting information as the value that a sophisticated investor can extract from firm-specific financial statement data when maximizing his expected utility from holding a portfolio of U.S equities As such, our paper directly studies the extent to which *UNC Chapel Hill We appreciate the comments of Sanjeev Bhojraj, Jim Ohlson, and workshop participants at UNC Chapel Hill and the 2009 JAAF Conference, especially Suresh Govindaraj (discussant) We are also very grateful to Michael Brandt, Pedro Santa-Clara, and Ross Valkanov for sharing important parts of their MATLAB code with us Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 JOURNAL OF ACCOUNTING, AUDITING & FINANCE portfolio performance may be improved by using accounting information Under our definition, accounting data are economically important if they lead the investor to materially tilt his portfolio weights away from the value-weighted market and toward under- or overweightings that are predictably related to firm-specific, publicly available financial statement data items Relatively few papers in finance and accounting adopt a portfolio-tilt approach to evaluating the significance of investment signals And among those that do, the dominant method has been to tilt portfolio weights toward factormimicking portfolios, not firm-specific characteristics.1 The main reason for this is that incorporating firm characteristics into traditional mean-variance analysis (Markowitz [1952]) requires modeling every firm’s expected return, variance, and covariances as a function of those characteristics Not only is this an intimidating task given the number of elements involved, but also Markowitz portfolio solutions are notoriously unstable and often yield error-maximizing extreme weights (Michaud [1989]) Although these problems can to some extent be mitigated by either applying shrinkage techniques to parameter estimates, imposing a factor structure onto returns, or imposing pragmatic constraints on the magnitude of permissible portfolio weights, they are severe enough to dissuade all but large sophisticated quantitative asset managers from optimizing their portfolios using firm-specific characteristics The notable exception to this discouraging picture is the recent work of Brandt, Santa-Clara, and Valkanov (2009, hereafter BSCV) BSCV develop a simple yet ingenious parametric portfolio policy (PPP) technique that directly models stocks’ portfolio weights as a linear function of firm characteristics They then estimate the policy’s few parameters by maximizing the average utility that an optimizing investor would have realized had she implemented the policy over the sample period The characteristics with which they illustrate their PPP method are firm size, book-to-market, and momentum BSCV’s results indicate that book-tomarket and momentum are highly significant in explaining portfolio tilt weights Grinold (1992) finds that in four out of five countries investors can substantially improve the in-sample Sharpe ratio of their tangency equity portfolio by tilting toward volatility, momentum, size, and value factor portfolios Similarly, Haugen and Baker (1996) use expected returns estimated from non-CAPM multifactor models to construct optimized portfolios that in-sample dominate the meanvariance locations of capitalization-weighted market index for the United States, United Kingdom, France, Germany, and Japan Korkie and Turtle (2002) investigate the extent to which dollar-neutral ‘‘overlay’’ assets created out of Fama-French market capitalization and value portfolios can expand the in-sample efficient frontier Kothari and Shanken (2002) explore the empirical limitations of the CAPM by estimating the degree to which investors should tilt their portfolios away from the market index to exploit the apparently anomalous returns in value, momentum, and size-based trading strategies Mashruwala, Rajgopal, and Shevlin (2006) apply Kothari and Shanken’s approach to estimating the optimal in-sample tilt toward a long-short accrual hedge portfolio strategy They find that an investor would invest approximately 50 percent in the value-weighted index and 50 percent in the long-short accrual hedge portfolio, thereby earning an in-sample additional 3.2 percent per year on their overall portfolio of U.S equity Finally, Hirshleifer, Hou, and Teoh (2006) find that an accrual factor-mimicking portfolio materially increases the Sharpe ratio of the ex post mean-variance tangency portfolio facing investors Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 THE IMPORTANCE OF ACCOUNTING INFORMATION and produce a certainty-equivalent out-of-sample return of 5.4 percent per year incremental to that of the value-weighted market A key source of the benefits realized from BSCV’s PPP method is its ability to simultaneously capture the relations between firm-specific characteristics and expected returns, variances, and covariances because all these moments affect the distribution of the optimized portfolio’s returns The PPP method allows a given firm-specific financial statement data item to tilt a stock in two ways: by generating alpha (long or short) or by reducing portfolio risk In contrast, a conventional dollar-neutral long-short hedge portfolio constructed to test the efficiency of the stock market with respect to the same financial statement data item does not take portfolio risk reduction into account, especially if the investor is optimizing over multiple financial statement data items The PPP method is not equivalent to testing whether a firm-specific accounting-based or price-based signal is cross-sectionally related to the conditional moments of stock returns This is because a signal may be correlated with the first and second moments of stock returns in offsetting ways with the result that the investor’s conditionally optimal portfolio weights are unrelated to the signal The goal of this paper is to use BSCV’s PPP method to model stocks’ portfolio weights as a linear function of three simple, publicly available firm-specific accounting-based signals: annual accruals, annual change in earnings, and annual asset growth We compare and contrast the total and incremental importance of the accounting-based signals with those of an illustrative set of price-based signals by modeling the weights as a linear function of not just accruals, change in earnings, and asset growth, but also firm size, book-to-market, and momentum Using monthly return data on U.S stocks between 1965 and 2008, we find that accruals, change in earnings, and asset growth are economically important in that the portfolio tilt weights they generate yield a 34-year (1975–2008) out-of-sample pre-transaction-costs annual information ratio of 1.9 This compares favorably to the information ratio of 1.5 achieved by firm size, book-to-market, and momentum When investors optimize over all six characteristics, the information ratio increases to 2.0 Such information ratios rank in the top decile of before-fee information ratios according to statistics reported by Grinold and Kahn (2000, Table 5.1) The delevered hedge portion of the portfolio optimized over all six firm characteristics yields out-of-sample pre-transactions-costs return that has a mean raw (alpha) return of 11.3 percent (13.2%) per year, a standard deviation (residual standard deviation) of 5.9 percent (6.2%), and a Sharpe (1994) ratio (information ratio) of 1.9 (2.1) The long and the short sides of this hedge portfolio contribute about equally to the size of the overall mean hedge return When short sales are disallowed, however, we find that the performance of the optimal portfolios suffers considerably The out-of-sample annual information ratio of the price-based characteristics optimal portfolio drops from 1.5 to 0.8, and that of the accounting-based characteristics optimal portfolio plummets from 1.7 to 0.1 This indicates that being able to short sell is particularly vital for investors looking to extract value from firm-specific accounting information Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 JOURNAL OF ACCOUNTING, AUDITING & FINANCE Consistent with its sizeable and stable performance, we find that the (delevered) hedge portion of the accounting-based optimal portfolio performed well during the severe bear market of 2008 The accounting-based hedge portfolio earned 12 percent during 2008 versus 3.4 percent for the price-based hedge and À38 percent for the value-weighted market We infer from these results that accountingbased firm characteristics are economically important to investors over and above their general long-run return properties, because when they are used optimally in portfolio construction they—unlike many hedge funds—yield returns that appear to remain truly hedged even in the face of severe market declines From our analyses we conclude if the economic importance of accounting information is defined by the value that sophisticated investors can extract from financial statements when maximizing the utility they expect from a portfolio of U.S equities, then accounting information is indeed important because optimally tilting a value-weighted portfolio toward firms with certain characteristics produces returns that are both high and stable Our evidence also suggests that accountingbased characteristics seem particularly valuable in periods when severe negative shocks are experienced in the stock market as a whole We show that large returns may be available to sophisticated investors as long as they can short sell from exploiting the illustrative financial statement signals that we study We also suspect that even better post-transactions-costs returns can be earned, for example, by trading much closer to when financial statements are first made public, directly including transactions costs in the portfolio optimization, and using a larger set and more diverse of accounting-based signals The remainder of the paper is structured as follows In Section 2, we explain the intuition and algebra of BSCV’s PPP method, and discuss its strengths and limitations We describe the data we use and our implementation timeline in Section 3, and our empirical results in Section In Section 5, we outline some caveats to our analyses We conclude in Section The Parametric Portfolio Policy Method of Brandt, Santa-Clara, and Valkanov (2009) 2.1 Basic Structure of the Parametric Portfolio Policy Method The PPP method begins by assuming that at every date t the investor choot ses a set of portfolio weights fwit gNi¼1 over a set of stocks Nt to maximize the conditional expected utility of that portfolio’s one-period ahead return rp,tþ1:2 max t fwit gNi¼1 " Et ½uðrp;tþ1 ފ ¼ Et u Nt X !# wit ri;tþ1 ð1Þ i¼1 BSCV emphasize that the PPP method can accommodate all kinds of objective functions, not just that chosen per eq (1), such as maximization of the portfolio’s Sharpe or information ratio Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 THE IMPORTANCE OF ACCOUNTING INFORMATION What distinguishes the PPP approach from a conventional mean-variance analysis (Markowitz [1952]) is the proposition that the investor’s optimal portfolio weights can be parameterized as a function of a vector of stocks’ characteristics xit observed at t: wit ¼ f ðxit ; hÞ ð2Þ As detailed by BSCV, the main conceptual advantages of the PPP approach are that it avoids the difficult task of modeling the joint distribution of returns and characteristics; it dramatically reduces the dimensionality of the optimization problem; it simultaneously takes into account the relations between firm characteristics and all return moments; and it accommodates all sorts of investor objective functions The PPP method also has several practical advantages It is easy and fast to implement in terms of computer run-time; it produces out-of-sample results that typically are only slightly worse than their in-sample counterparts because the parsimonious number of parameters involved reduces the risk of overfitting; it does not typically produce extreme portfolio weights; and it can be modified readily to allow for short-sale constraints, transactions costs, and nonlinear parameterizations of eq (2), such as interactions between firm characteristics or conditioning on macroeconomic variables Our focus in this study is on two illustrative sets of firm-specific attributes The first consists of three accounting-based characteristics (ABCs) that are entirely and only contained in a firm’s financial statements: accruals, change in earnings, and asset growth The second set consists of three price-based characteristics (PBCs) that are partially or fully defined using a firm’s stock price or stock return: market capitalization, book-to-market, and momentum Following BSCV, we adopt a linear specification for the portfolio weight function: "it þ wit ¼ w T^ h xit Nt ð3Þ "it is the weight of stock i in a benchmark portfolio, which we take to be where w the value-weighted equity market VW, y is a vector of coefficients, and x^it are the characteristics of stock i after they have been cross-sectionally standardized at t Eq (3) also expresses the idea of active portfolio management since T x ^it describe the tilts—whether positive or negative—of the optimal portfolio weights from VW Moreover, because the tilt weights must sum to zero, the difference between the return on the optimal portfolio rp,tþ1 and the return rm,tþ1 on VW represents the return to a long-short levered hedge portfolio rlevh,tþ1: rp;tþ1 ¼ Nt X i¼1 "it ri;tþ1 þ w Nt  X i¼1 Nt hTx^it  ri;tþ1 ¼ rm;tþ1 þ rlevh;tþ1 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 ð4Þ JOURNAL OF ACCOUNTING, AUDITING & FINANCE To calibrate hedge returns across the different optimal portfolios generated by applying the PPP method to different sets of firm characteristics, we delever rlevh,tþ1 by separately calculating the returns on the long and short sides of the hedge, rhLong and rhShort : Long Short þ rh;tþ1 Þ rlevh;tþ1 ¼ opt levh;tþ1 ðrh;tþ1 ð5Þ where Long ¼ rht NLt X wLjt rjt and Short rht ¼ j¼1 wjt wLjt ¼ NL Pt wjt NSt X wSkt rkt ð6Þ k¼1 and j¼1 wkt NSt P wkt wSkt ¼ ð7Þ k¼1 and NLt (NSt) is the number of firms at time t with a positive (negative) tilt weight T x ^it > ( T x ^it < ) Then the delevered hedge return on the optimal portfolio rh,tþ1 is as follows: Long Short þ rht rh;tþ1 ¼ rht ð8Þ and the leverage of the optimal portfolio is as follows: opt levh;tþ1 ¼ rlevh;tþ1 rh;tþ1 ð9Þ As emphasized by BSCV, a key feature of the parameterization in eq (2) is that y is constant across assets and through time This means that the conditional optimization with respect to wit in eq (1) can be rewritten as an unconditional optimization with respect to y Thus, for a given utility function, the optimization problem translates to empirically estimating y in eq (10) over the sample period: !  Nt  TÀ1 X max X T^ "i;t þ h xit ri;tþ1 u w Nt h T t¼0 i¼1 ð10Þ Following BSCV, we assume the investor’s utility function is such that he has constant relative risk-aversion preferences over wealth, where following BSCV we set g ¼ 5: uðrp;tþ1 Þ ¼ ð1 þ rp;tþ1 Þ1Àc 1Àc Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 ð11Þ THE IMPORTANCE OF ACCOUNTING INFORMATION Data and Implementation Timeline 3.1 Variable Definitions and Sample Selection Criteria We collect financial statement data from the Compustat annual industrial file; monthly stock returns, including delisting returns, from Center for Research in Security Prices (CRSP) monthly files; and the one-month Treasury bill rate from the Fama-French factor data set at Wharton Research Data Services (WRDS) We select all Compustat variables from fiscal year 1964 (and lagged variables from 1963) through fiscal year 2007 together with CRSP data from January 1965 through December 2008 Following BSCV, we restrict the investor’s opportunity set to U.S stocks and not include the risk-free asset because a first-order approximation including the risk-free asset affects only the leverage of the optimized portfolio Panel A of Table defines the illustrative firm characteristics that we employ to assess the economic importance of accounting information using the PPP method Firm size is defined as the market value of common equity (MVE) at the firm’s fiscal year end (Banz [1981]; Reinganum [1981]) Book-to-market (BTM) is the fiscal year-end book value of common equity scaled by MVE (Stattman [1980]; Rosenberg, Reid, & Lanstein [1985]) Momentum (MOM) is taken to be the cumulative raw return for the twelve months ending four months after the most recent fiscal year-end (Jegadeesh [1990]; Jegadeesh & Titman [1993]) When the statement of cash flows is available, annual accruals (ACC) are net income less operating cash flow scaled by average total assets; otherwise per Sloan (1996), we set ACC ¼ Dcurrent assets – Dcash – Dcurrent liabilities – Ddebt in current liabilities – Dtaxes payable – Ddepreciation all scaled by average total assets (where if any one of the balance sheet–based accrual components is missing, we set it to zero) The change in annual earnings (UE) is simply earnings in the most recent fiscal year less earnings one-year prior, scaled by average total assets (Ball & Brown [1968]; Foster, Olsen, & Shevlin [1984]) Lastly, asset growth (AGR) is defined to be the natural log of total assets at the end of the most recent fiscal year less the natural log of total assets one year earlier (Cooper, Gulen, & Schill [2008, 2009]) Panel B of Table reports the number of observations we start with and the number that are eliminated as we require firms first to have all the price-based data items, and second all the accounting-based data items Each step removes approximately percent of the initial total data set that requires only the availability of monthly stock returns We also follow BSCV by deleting the smallest 20 percent of firms as measured by MVE since such firms tend to have low liquidity, high bid-ask spreads, and disproportionately high transactions costs As shown by the graph below Table 1, Panel B, the final number of firms varies greatly by year, rising from a low of only 276 in 1965 to a peak of 5,615 in 1998 The average number of firms per year is 3,373 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 JOURNAL OF ACCOUNTING, AUDITING & FINANCE TABLE Variable Definitions and Sample Selection Criteria Panel A: Definitions of accounting-based firm characteristics (ABCs) and price-based firm characteristics (FBCs) used in the Brandt, Santa-Clara, and Valkanov (BSCV) portfolio optimizations Firm size (MVE) ¼ fiscal year end market value of common equity Book-to-market (BTM) ¼ book value of common equity / MVE Momentum (MOM) ¼ cumulative raw return for the twelve months ending four months after the most recent fiscal year end Accruals (ACC) ¼ net income – operating cash flow scaled by average total assets if operating cash flow is available, otherwise ACC ¼ Dcurrent assets – Dcash – Dcurrent liabilities – Ddebt in current liabilities – taxes payable – depreciation, all scaled by average total assets If any of the aforementioned components is missing, we set it to zero Change in earnings (UE) ¼ change in net income scaled by average total assets Asset growth (AGR) ¼ ln [1 þ total assets] – ln [1 þ lagged total assets] Panel B: Sample selection criteria and data restrictions No of Monthly Obs Initial set of unrestricted monthly stock returns After requiring sufficient data to compute a firm’s book-to-market, market capitalization, and twelve-month stock return momentum After also requiring sufficient data to compute a firm’s accruals, change in earnings, and asset growth After deleting the smallest 20 percent of stocks Percent 2,642,298 2,392,504 100% 91% 2,168,745 82% 1,735,192 66% Note: This table reports the definitions of data items employed in the estimations (Panel A) and the restrictions imposed in arriving at the sample used to estimate the parameters in our application of Brandt, Santa-Clara, and Valkanov’s (2009) linear parametric portfolio policy method (Panel B) Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 THE EFFECT OF RESEARCH AND DEVELOPMENT (R&D) 135 TABLE Cross-Sectional Regression Tests of the Level of Future Operating Performance Panel A: Regression of mean adjusted net income (NET_INC)a Pred Sign Model Model COUNT þ CITATION þ COUNT*R&D þ CITATION*R&D þ COUNT* CITATION þ R&D þ CAPEX þ ADVT þ PAST_NI þ SIZE ? LEV ? AGE ? À0.012 (À0.09) 0.031 (0.68) 0.072*** (3.07) 0.215*** (7.31) 0.560*** (38.25) 0.012*** (10.42) À0.028** (À2.20) 0.032*** (15.13) À0.244*** (À3.17) Present 0.163 20,391 Intercept Industry Dummies Adj R2 N Model Model ** À0.015** (À2.09) 0.008*** (3.20) À0.009 (À1.03) 0.024 (0.77) 0.003*** 0.009*** (5.40) À0.027 (À2.27) 0.010*** (5.64) 0.023 (0.50) 0.071*** (3.02) 0.214*** (7.26) 0.559*** (38.23) 0.011*** (9.59) À0.029** (À2.22) 0.031*** (14.60) À0.241*** (À3.09) Present 0.164 20,391 0.024 (0.53) 0.072*** (3.07) 0.214*** (7.27) 0.559*** (38.23) 0.011*** (9.74) À0.028** (À2.20) 0.031*** (14.56) À0.242*** (À3.11) Present 0.164 20,391 (3.15) 0.018 (0.37) 0.073*** (3.12) 0.213*** (7.25) 0.560*** (38.24) 0.012*** (9.82) À0.028** (À2.19) 0.031*** (14.46) À0.242*** (À3.12) Present 0.165 20,391 Panel B: Regression of mean adjusted cash flow from operations (CFO)b Pred Sign COUNT þ CITATION þ COUNT*R&D þ CITATION*R&D þ Model Model Model * 0.017 (1.03) 0.015*** (6.55) À0.022 (À1.70) 0.015*** (6.54) Model À0.015*** (À3.01) 0.013*** (3.92) 0.089 (0.22) 0.030 (0.75) (Continued) Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 136 JOURNAL OF ACCOUNTING, AUDITING & FINANCE TABLE (Continued) COUNT*CITATION þ R&D þ CAPEX þ ADVT þ PAST_CFO þ SIZE ? LEV ? AGE ? Intercept Industry Dummies Adj R2 N À0.140** (À2.34) 0.161*** (4.95) 0.171*** (4.11) 0.609*** (31.08) 0.014*** (9.50) À0.061*** (À3.41) 0.040*** (14.90) À0.279*** (À2.71) Present 0.194 20,391 À0.151** (À2.53) 0.160*** (4.94) 0.170*** (4.09) 0.609*** (31.15) 0.013*** (8.61) À0.060*** (À3.41) 0.039*** (14.29) À0.276*** (À2.62) Present 0.195 20,391 À0.150** (À2.51) 0.161*** (4.97) 0.170*** (4.09) 0.608*** (31.15) 0.013*** (8.66) À0.060*** (À3.40) 0.039*** (14.27) À0.277*** (À2.63) Present 0.195 20,391 0.031*** (3.06) À0.165*** (À2.63) 0.161*** (4.95) 0.170*** (4.08) 0.608*** (31.13) 0.013*** (8.73) À0.060*** (À3.38) 0.039*** (14.18) À0.277*** (À2.64) Present 0.195 20,391 Note: See Appendix A for variable definitions Robust standard errors are clustered at the firm level (e.g., Rogers [1993]; Petersen [2009]) *** ** , , and * Denote significance at 1, 5, and 10 percent levels, respectively a The sample consists of 20,391 firm-year observations with available data over 1972–2000 The dependent variable (NET_INC) is defined as adjusted net income, deflated by lagged market value of equity and averaged over years [tþ1, tþ5] The table reports results of the following regression: NET INC ¼  þ  COUNT þ  CITATION þ  COUNT à R&D þ  CITATION à R&D þ  COUNT à CITATION þ  R&D þ  CAPEX þ  ADVT þ  PAST NI þ  10 SIZE þ  11 LEV þ  12 AGE þ error: b The sample consists of 20,391 firm-year observations with available data over 1972–2000 The dependent variable (CFO) is defined as adjusted cash flow from operations, deflated by lagged market value of equity and averaged over years [tþ1, tþ5] The table reports results of the following regression: CFO ¼  þ  COUNT þ  CITATION þ  COUNT à R&D þ  CITATION à R&D þ  COUNT à CITATION þ  R&D þ  CAPEX þ  ADVT þ  PAST CFO þ 10 SIZE þ  11 LEV þ  12 AGE þ error: Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 THE EFFECT OF RESEARCH AND DEVELOPMENT (R&D) 137 on the level of future operating performance Subsequently, in Model 3, we include both variables to assess their incremental explanatory power Finally, in Model we introduce some interactive terms between the patent variables and R&D In Panel A of Table 3, where the dependent variable is NET_INC, the results for Model reveal that COUNT is not significantly associated with mean future net income (coefficient ¼ À0.012, robust t-statistic ¼ À0.09) On the other hand, in Model CITATION is significantly positively associated with NET_INC (coefficient ¼ 0.009, t-statistic ¼ 5.40) These results provide evidence that possessing a large number of patents does not help future operating performance, but having highly cited patents does When both patent variables are included in Model 3, CITATION remains significantly positive, while COUNT is negative and significant (coefficient À0.027, robust t-statistic À2.27) With regard to the other variables in the model, CAPEX and ADVT have significantly positive coefficients, indicating that firms investing in tangible assets and marketing activities reap the benefits of those investments SIZE and AGE have significantly positive coefficients, an indication that larger and older firms tend to exhibit superior performance, while lagged net income, PAST_NI, has a significantly positive effect in all models, consistent with persistence in operating performance Model includes all variables in Models through as well as interactive terms between COUNT, CITATION, and R&D All main effects in the model retain their sign and significance Among the interactive variables, COUNT*CITATION is significantly positive, indicating that firms that have a larger number of highly cited patents exhibit superior net income over the next five years Other interactive terms are insignificant Lastly, the adjusted R2 is over 16 percent for Models to 4, implying a reasonable fit, although we note that the incremental explanatory power of the patent-related variables (COUNT and CITATION) is not very high Panel B of Table reports results where future operating performance is measured as the average level of adjusted cash flows from operations deflated by lagged market value of equity over years (tþ1, tþ5) An advantage of this measure is that it is relatively free from the effect of managers’ accounting choices, and thus constitutes a potentially more precise measure of future operating performance when compared with NET_INC Consistent with our predictions, the results for Models through reveal that CITATION has a strong positive association with future cash flow performance, while COUNT is weakly negative The other variables have similar coefficients as in the earlier NET_INC model, except for R&D that has a significantly negative coefficient, an indication that firms with high R&D experience lower future operating cash flows Turning to Model 4, COUNT has a significant negative coefficient while CITATION continues to exhibit a significantly positive association with future operating cash flows Also, the interactive variable COUNT*CITATION is significantly positive as in the case of net income in Panel A of Table Again, this is evidence that firms with a larger patent portfolio of more highly cited patents exhibit superior operating cash flows Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 138 JOURNAL OF ACCOUNTING, AUDITING & FINANCE Overall, the results in Table support our prediction that highly innovative firms enjoy better future accounting operating performance In particular, we find that it is not sufficient to have a large patent stock Rather, it is the quality of the patents possessed by a firm, as evidenced by citations from other patents that is a key factor in explaining future operating performance 4.2 The Volatility of Future Operating Performance Table presents results of cross-sectional regression tests of future operating performance volatility As shown by the results for Model in Panel A, COUNT itself does not exhibit a significant association with the standard deviation of NET_INC (coefficient ¼ À0.006, robust t-statistic ¼ À0.50) However, in Model CITATION has a significantly negative coefficient of À0.011 (robust t-statistic ¼ À7.47) When both variables are included together in Model 3, COUNT has a weakly positive coefficient while CITATION retains its significant negative coefficient These results provide evidence that is consistent with firms possessing highly cited patents exhibiting less volatile future operating performance Turning to the other variables in Model 3, R&D and CAPEX have significantly positive coefficients, which is consistent with findings in prior studies such as Kothari, Laguerre, and Leone (2002) who posit that future benefits from R&D outlays are more uncertain than those of capital expenditures On the other hand, ADVT has a significant negative coefficient, an indication that firms with high levels of advertising expenditures have a less volatile stream of future net income As expected, the past volatility of net income, STDEV_PAST_NI, is highly positively associated with the volatility of realized future net income suggesting that operating performance volatility is persistent Finally, SIZE and AGE have negative coefficients, implying that larger and older firms have lower income volatility In Model 4, we regress the standard deviation of NET_INC on interactive terms in addition to all variables included in Model We find that COUNT and CITATION continue to have coefficients as before and that the interactive term COUNT*CITATION has a significant negative coefficient, suggesting that firms with a higher number of well-cited patents experience more stable operating performance In addition, the interactive term CITATION*R&D has a negative coefficient (À0.045, t-statistic À1.84), which implies that firms that are able to convert their R&D expenditures into highly cited patents are able to enjoy more stable performance Panel B of Table reports results of cross-sectional regression tests of the standard deviation of cash flows from operations deflated by lagged market value of equity regressed on the patent measures and the other variables In Panel B of Table 3, we found that patent citations are significantly positively associated with the level of future operating cash flows Our interest in Table is whether the patent variables are negatively associated with the volatility of future operating cash flows Turning directly to Model 4, COUNT is significantly positively associated with the standard deviation of future CFO while CITATION is significantly Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 THE EFFECT OF RESEARCH AND DEVELOPMENT (R&D) 139 TABLE Cross-Sectional Regression Tests of the Volatility of Future Operating Performance Panel A: Regression of the standard deviation of adjusted net income: STDEV(NET_INC)a Pred Sign Model Model Model Model COUNT À CITATION À COUNT*R&D À CITATION*R&D À COUNT*CITATION À R&D þ CAPEX þ ADVT þ STDEV(PAST_NI) þ SIZE À LEV þ AGE ? Intercept Industry Dummies Adj R2 N À0.011*** (À7.47) 0.024* (1.88) À0.012*** (À7.67) 0.191*** (5.51) 0.057*** (3.28) À0.051** (À2.25) 0.280*** (19.00) À0.007*** (À6.88) À0.001 (À0.05) À0.041*** (À21.66) 0.184** (2.28) Present 0.127 20,391 0.190*** (5.48) 0.056*** (3.22) À0.051** (À2.52) 0.280*** (18.97) À0.007*** (À7.03) À0.001 (À0.07) À0.041*** (À21.62) 0.186** (2.30) Present 0.127 20,391 À0.006 (À0.50) 0.181*** (5.25) 0.056*** (3.21) À0.052*** (À2.59) 0.282*** (19.13) À0.008*** (À7.91) À0.001 (À0.10) À0.042*** (À22.25) 0.187** (2.37) Present 0.124 20,391 Panel B: Regression of standard deviation of adjusted cash flow: STDEV(CFO)b Pred Sign Model Model Model COUNT À CITATION À COUNT*R&D À CITATION*R&D À À0.012 (À0.88) À0.012*** (À7.79) 0.020 (1.30) À0.013*** (À7.93) 107** (2.01) À0.008*** (À3.88) 0.190 (0.71) À0.045* (À1.84) À0.024** (À2.19) 0.209*** (5.55) 0.056*** (3.22) À0.051** (À2.51) 0.281*** (19.01) À0.007*** (À7.12) À0.001 (À0.11) À0.041*** (À21.54) 0.185** (2.31) Present 0.128 20,391 Model 0.122** (2.10) À0.010*** (À4.23) 0.093 (0.28) À0.037 (À1.37) (Continued) Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 140 JOURNAL OF ACCOUNTING, AUDITING & FINANCE TABLE (Continued) COUNT*CITATION À R&D þ CAPEX þ ADVT þ STDEV(PAST_CFO) þ SIZE À LEV þ AGE ? Intercept Industry Dummies Adj R2 N 0.206*** (5.26) 0.120*** (5.84) À0.058** (À2.49) 0.334*** (20.84) À0.008*** (À7.59) 0.041*** (3.45) À0.043*** (À21.33) 0.210** (2.40) Present 0.149 20,391 0.216*** (5.49) 0.121*** (5.89) À0.057** (À2.44) 0.332*** (20.82) À0.007*** (À6.62) 0.041*** (3.50) À0.042*** (À20.70) 0.207** (2.31) Present 0.151 20,391 0.215*** (5.47) 0.120*** (5.84) À0.057** (À2.45) 0.331*** (20.79) À0.008*** (À6.72) 0.041*** (3.49) À0.042*** (À20.67) 0.209** (2.32) Present 0.152 20,391 À0.027** (À2.30) 0.232*** (5.44) 0.120*** (5.82) À0.056** (À2.43) 0.332*** (20.78) À0.008*** (À6.80) 0.041*** (3.47) À0.042*** (À20.57) 0.208** (2.33) Present 0.152 20,391 Note: See Appendix A for variable definitions Robust standard errors are clustered at the firm level (e.g., Rogers [1993]; Petersen [2009]) *** ** , , and * Denote significance at 1, 5, and 10 percent levels, respectively a The sample consists of 20,391 firm-year observations with available data over 1972–2000 The dependent variable is the standard deviation of net income calculated over years [tþ1, tþ5] The table reports results of the following regression: STDEV ðNET INCÞ ¼  þ  COUNT þ  CITATION þ  COUNT à R&D þ 4 CITATION à R&D þ  COUNT à CITATION þ  R&D þ  CAPEX þ  ADVT þ 9 STDEV ðPAST NIÞ þ  10 SIZE þ  11 LEV þ 12 AGE þ error: b The sample consists of 20,391 firm-year observations with available data over 1972–2000 The dependent variable is the standard deviation of cash flow from operations (CFO), calculated over years [tþ1, tþ5] The table reports results of the following regression: STD DEV ðCFOÞ ¼  þ  COUNT þ  CITATION þ  COUNT à R&D þ  CITATION à R&D þ  COUNT ÃCITATION þ  R&D þ 7 CAPEX þ  ADVT þ  STDEV ðPAST CFOÞ þ  10 SIZE þ  11 LEV þ 12 AGE þ error: negatively associated with future cash flow volatility (coefficient¼ À0.010, tstatistic¼ À4.23) For the interactive terms in the model, we find that COUNT*CITATION has a significantly negative coefficient This evidence suggests that firms that have a large number of patents exhibit more variability in future cash flows, although firms with more highly cited patents experience more stable (less volatile) future cash flows For the remaining variables, volatility of past CFO has a Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 THE EFFECT OF RESEARCH AND DEVELOPMENT (R&D) 141 positive association with the volatility of future CFO, and cash flow volatility is positively (negatively) associated with LEV (SIZE and AGE) In summary, the evidence in Table is consistent with our prediction that firms possessing highly cited patents experience less volatile (i.e., more stable) future operating performance In particular, we find that the interaction between a larger number of patents and higher quality patents is associated with more stable future firm performance.17 Conclusion Empirical accounting research on the informational role of R&D largely focuses on the treatment of R&D for financial reporting purposes and the value relevance of R&D Prior studies also find a positive relationship between R&D and the uncertainty of future benefits, particularly in R&D-intensive industries While R&D expense captures one of the inputs into the innovation process, it does not capture any of the outputs of that process that serve as drivers of the firm’s production function Moreover, because firms vary in their ability to convert R&D spending into real benefits, the assumption in prior studies that the marginal benefit of R&D is constant across firms is problematic In this study, we relax this assumption and examine the relation between both the inputs and outputs of innovation with the level and volatility of future operating performance We use R&D outlays as a proxy for innovation effort (i.e., inputs) and patent counts and citations as proxies for the outcome of innovation (i.e., outputs) An advantage of our study’s design is that it allows for a comprehensive examination of how innovation translates into future operating performance Using a sample of 20,391 firm-year observations with nonzero R&D outlays over the 1972–2000 time period, we find that (1) the mean level of future operating performance (net income or cash flows deflated by lagged market value) is positively associated with the quality of the patents a firm possesses as measured by the mean citation index across the firm’s patent portfolio and (2) the standard deviation of future operating performance is negatively associated with the 17 Some studies argue a firm’s operating performance, its decision to invest in R&D, and the outcome of its innovative activities such as patents are endogenously determined (see Klomp & Van Leeuwen [2001]; Loof & Heshmati [2006]) Thus, empirical studies of innovation and operating performance need to account for this potential endogeneity To ensure our results are robust, we implement a simultaneous equations estimation procedure that includes the following equations: OP_PERF ¼ a1 þ b1COUNT þ b2CITATION þ b3COUNT*R&D þ b4CITATION*R&D þ b5COUNT*CITATION þ b6R&D þ b7CAPEX þ b8ADVT þ b9PAST_OP_PERF þ b10SIZE þ b11LEV þ b12AGE þ error, R&D ¼ a2 þ g1Mean(CFO) þ g2SIZE þ g3LEV þ g4AGE þ error, CITATION ¼ a3 þ f1R&D þ f2Mean(CFO) þ f3SIZE þ f4LEV þ f5AGE þ error The results remain consistent with our predictions In particular, after controlling for the potential endogeneity of operating performance, R&D expense, and innovation output, firms possessing more valuable patents exhibit higher levels of and lower variability of future operating performance Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 142 JOURNAL OF ACCOUNTING, AUDITING & FINANCE quality of the firm’s patents, and this association is stronger for firms with higher R&D expenditures Our main tests are robust to accounting for the endogeneity between operating performance, R&D outlays, and patent activity, using a simultaneous equations approach Our main findings suggest that more innovative firms have higher levels of and lower uncertainty of operating performance Such results partly explain the results in prior research that highly innovative firms enjoy higher market capitalizations (Trajtenberg [1990]; Harhoff et al [1999]; Hall, Jaffe, & Trajtenberg [2005]) In particular, our results suggest that this is not only due to the real option value embedded in patents, but also to the realized superior future operating performance exhibited by innovative firms By shedding light on the relation between innovation and operating performance, our study contributes to the accounting literature on the informational role of R&D We establish a link between firm-level innovation and the level and volatility of future operating performance, but we not decompose operating performance to examine the mechanisms through which innovation affects operating performance For example, it is possible that some patents lead to an increase in revenues through higher product sales, royalties, or licensing fees, while other patents lead to cost savings within the firm An interesting extension of our study would be to examine more closely how patents and innovation affect future operating performance APPENDIX A Variable Definition ADVT (Advertising expense deflated by lagged market value of equity) AGE (firm age) CAPEX (Capital expenditures deflated by lagged market value of equity) CFO (Mean of adjusted cash flows from operations deflated by lagged market value of equity over years tþ1 through tþ5) CITATION (patent citations) COUNT (patent stock) LEV (leverage) DATA45t/(DATA25*DATA199)t-1 The number of years since the firm first appeared on the Compustat or Center for Research in Security Prices (CRSP) database (whichever is earlier) The natural logarithm of AGE in used in the regressions DATA128t/(DATA25*DATA199)t-1 (DATA308 þ DATA45 þ DATA46)t/ (DATA25*DATA199)t-1 where available, otherwise (DATA178 – Total Accruals þ DATA45 þ DATA46)t / (DATA25* DATA199)t-1 where Total Accruals are calculated as in Sloan (1996) The mean and standard deviation of CFO are calculated over years [tþ1, tþ5] The average across all patents issued to firm j in the previous five years of: (The number of citations received in year t by patent i of firm j / the mean number of citations received up to year t by all patents in the same technological category that were issued in the same year as patent i of firm j) The number of patents granted to firm j in the five years up to year t In regression tests, we scale COUNT by 1,000 (DATA9 þ DATA34)t/(DATA9 þ DATA34 þ DATA25*DATA199)t Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 THE EFFECT OF RESEARCH AND DEVELOPMENT (R&D) NET_INC (Mean of adjusted net income deflated by lagged market value of equity over years tþ1 through tþ5) PAST_CFO (Adjusted cash flows from operations deflated by lagged market value of equity) PAST_NI (Adjusted net income deflated by lagged market value of equity) R&D (Research and development expense deflated by lagged market value of equity) SIZE (market value of equity) Z (Altman’s Z-score) 143 ((DATA172 þ (DATA45þDATA46)*(1 – 0.35))t / (DATA25*DATA199)t-1 The mean and standard deviation of NET_INC are calculated over years [tþ1, tþ5] (DATA308 þ DATA45 þ DATA46)t/ (DATA25*DATA199)t-1 where available, otherwise (DATA178 – Total Accruals þ DATA45 þ DATA46)t / (DATA25* DATA199)t-1 where Total Accruals are calculated as in Sloan (1996) ((DATA172 þ (DATA45þDATA46)*(1 – 0.35))t / (DATA25*DATA199)t-1 (DATA46/DATA25)t/(DATA25*DATA199)t-1 (DATA25*DATA199)t The natural logarithm of SIZE in used in regressions Z ¼ 1.2*(DATA4 – DATA1)t – (DATA5 – DATA34)t/ DATA6t þ 1.4*DATA36t/DATA6t þ 3.3*(DATA13 – DATA14)t/DATA6t þ 0.6*(DATA25*DATA199)t/ (DATA6 – DATA60)t þ DATA12t/DATA6t REFERENCES Aboody, D., and B Lev 1998 ‘‘The Value Relevance of Intangibles: The Case of Software Capitalization.’’ Journal of Accounting Research 36 (Supplement): 161–191 Aghion, P., J Van Reenen, and L Zingales 2009 ‘‘Innovation and Institutional Ownership.’’ NBER Working Paper 14769, National Bureau of Economic Research, Cambridge, MA Altman, E 1968 ‘‘Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy.’’ Journal of Finance 23 (4): 589–609 Amir, E., and B Lev 1996 ‘‘Value-Relevance of Nonfinancial Information: The Wireless Communications Industry.’’ Journal of Accounting and Economics 22: 3–30 Barth, M., D Cram, and K Nelson 2001 ‘‘Accruals and the Prediction of Future Cash Flows.’’ The Accounting Review 76: 27–58 Bartov, E., P Mohanram, and C Seethamraju 2002 ‘‘Valuation of Internet Stocks: An IPO Perspective.’’ Journal of Accounting Research 40 (2): 321–347 Behn, B., and R Riley 1999 ‘‘Using Nonfinancial Measures to Predict Financial Performance: The Case of U.S Airline Industry.’’ Journal of Accounting Auditing and Finance 14: 29–56 Chambers, D., R Jennings, and R Thompson 2003 ‘‘Managerial Discretion and Accounting for Research and Development Costs.’’ Journal of Accounting Auditing and Finance 18: 79–113 Cheng, Q 2003 ‘‘What Determines Residual Income?’’ The Accounting Review 80: 85–112 Ciftci, M., B Lev, and S Radhakrishnan 2009 ‘‘Is R&D Mispriced or Properly Risk-Adjusted?’’ SSRN eLibrary, online database Dechow, P 1994 ‘‘Accounting Earnings and Cash Flows as Measures of Firm Performance: The Role of Accounting Accruals.’’ Journal of Accounting and Economics 18: 3–42 Demers, E., and B Lev 2001 ‘‘A Rude Awakening: Internet Shakeout in 2000.’’ Review of Accounting Studies 6: 331–359 Deng, Z., and B Lev 2006 ‘‘In-Process R&D: To Capitalize or Expense?’’ Journal of Engineering and Technology Management 23: 18–32 Deng, Z., B Lev, and F Narin 1999 ‘‘Science and Technology as Predictors of Stock Performance.’’ Financial Analysts Journal 55: 20–32 Eberhart, A., W Maxwell, and A Siddique 2008 ‘‘A Reexamination of the Tradeoff between the Future Benefit and Riskiness of R&D Increases.’’ Journal of Accounting Research 46 (1): 27–52 Finger, C 1994 ‘‘The Ability of Earnings to Predict Future Earnings and Cash Flow.’’ Journal of Accounting Research 32 (2): 210–223 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 144 JOURNAL OF ACCOUNTING, AUDITING & FINANCE Griliches, Z 1990 ‘‘Patent Statistics as Economic Indicators: A Survey.’’ Journal of Economic Literature 28: 1661–1707 Gu, F 2005 ‘‘Innovation, Future Earnings, and Market Efficiency.’’ Journal of Accounting Auditing and Finance 20: 385–418 Gu, F., and B Lev 2004 ‘‘The Information Content of Royalty Income.’’ Accounting Horizons 18: 1–12 Hall, B H 1993 ‘‘The Stock Market Valuation of R&D Investment during the 1980s.’’ American Economic Review 83: 259–264 Hall, B H., A B Jaffe, and M Trajtenberg 2001 ‘‘The NBER Patent Citations Data File: Lessons, Insights and Methodological Tools.’’ NBER Working Paper 8498, National Bureau of Economic Research, Cambridge, MA Hall, B H., A B Jaffe, and M Trajtenberg 2005 ‘‘Market Value and Patent Citations.’’ RAND Journal of Economics 36: 16–38 Harhoff, D., F Narin, F M Scherer, and K Vopel 1999 ‘‘Citation Frequency and the Value of Patented Innovation.’’ Review of Economics and Statistics 81 (3): 511–515 Ittner, C., and D Larcker 1998 ‘‘Are Nonfinancial Measures Leading Indicators of Financial Performance? An Analysis of Customer Satisfaction.’’ Journal of Accounting Research 36 (Supplement): 1–36 Klomp, L., and G Van Leeuwen 2001 ‘‘Linking Innovation and Firm Performance: A New Approach.’’ International Journal of the Economics of Business (3): 343–364 Kothari, S P., T E Laguerre, and A J Leone 2002 ‘‘Capitalization versus Expensing: Evidence on the Uncertainty of Future Earnings from Capital Expenditure versus R&D Outlays.’’ Review of Accounting Studies 7: 355–382 Lev, B., and T Sougiannis 1996 ‘‘The Capitalization, Amortization and Value-Relevance of R&D.’’ Journal of Accounting and Economics 21: 107–138 Loof, H., and A Heshmati 2006 ‘‘On the Relationship between Innovation and Performance: A Sensitivity Analysis.’’ Economics of Innovation and New Technology 15 (5): 317–344 Matolcsy, Z P., and A Wyatt 2008 ‘‘The Association between Technological Conditions and the Market Value of Equity.’’ The Accounting Review 83: 479–518 Narin, F 1995 ‘‘Patents as Indicators for the Evaluation of Industrial Research Output.’’ Scientometrics 34: 489–496 Pakes, A 1986 ‘‘Patents as Options: Some Estimates of the Value of Holding European Patent Stocks.’’ Econometrica 55: 755–784 Petersen, M 2009 ‘‘Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches.’’ Review of Financial Studies 22: 435–480 Rajgopal, S., T Shevlin, and M Venkatachalam 2003 ‘‘Does the Stock Market Fully Appreciate the Implications of Leading Indicators for Future Earnings? Evidence from Order Backlog.’’ Review of Accounting Studies 8: 461–492 Rogers, W 1993 ‘‘Regression Standard Errors in Clustered Samples.’’ Stata Technical Bulletin 13: 19–23 Shi, C 2003 ‘‘On the Trade-Off between the Future Benefits and Riskiness of R&D: A Bondholders’ Perspective.’’ Journal of Accounting and Economics 35: 227–254 Sloan, R 1996 ‘‘Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about Future Earnings?’’ The Accounting Review 71: 289–315 Sougiannis, T 1994 ‘‘The Accounting Based Valuation of Corporate R&D.’’ The Accounting Review 69: 44–68 Trajtenberg, M 1989 ‘‘The Welfare Analysis of Product Innovation with an Application to Computer Tomography Scanners.’’ Journal of Political Economy 97: 445–479 Trajtenberg, M 1990 ‘‘A Penny for Your Quotes: Patent Citations and the Value of Innovations.’’ RAND Journal of Economics 21: 172–187 Trueman, B., M Wong, and X Zhang 2000 ‘‘The Eyeballs Have It: Searching for the Value in Internet Stocks.’’ Journal of Accounting Research 38 (Supplement): 137–163 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 Discussion of ‘‘The Effect of Research and Development (R&D) Inputs and Outputs on the Relation between the Uncertainty of Future Operating Performance and R&D Expenditures’’ THEODORE SOUGIANNIS* Pandit, Wasley, and Zach (2011) examine whether both research and development (R&D) input (R&D expenditures) and output measures (patent counts and citations) and their interaction associate with the level and variability of future earnings and operating cash flows The associations examined contribute to the literature because they help determine whether the relationship between firm-level innovation and operating performance is conditional on the success of a firm’s R&D efforts The conditioning is important because it can sort out more successful from less successful R&D firms In my discussion, I focus on the models employed and on the empirical results Some results are consistent with expectations and some other results are not I think that the mixed results are due to the models employed and I make some suggestions for model improvements Keywords: Earnings Prediction, Earnings Variability, Patent Citations, Patent Counts, R&D Expenditures Introduction Shail Pandit, Charles E Wasley, and Tzachi Zach (2011) contribute to the R&D accounting literature by investigating whether both R&D input and output measures and their interaction associate with the level and variability of future earnings and operating cash flows They point out that prior related studies not consider the interaction between R&D inputs and outputs and implicitly assume ‘‘constant marginal productivity of R&D in the cross-section, which is unlikely to be the case.’’ Firm-level R&D expenditure is the measure of R&D input (effort), while patent counts (quantity) and citations (quality) are the measures of R&D output The results show that firms with highly cited patents exhibit superior and *University of Illinois and Athens Laboratory of Business Administration (ALBA) 145 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 146 JOURNAL OF ACCOUNTING, AUDITING & FINANCE less volatile (i.e., more stable) net income and operating cash flows over the next five years However, the results not show the interaction between R&D inputs and outputs to be significantly associated with either the level or the variability of future earnings and operating cash flows The evaluation of R&D productivity is an important issue that has received significant attention in the literature The associations that Pandit, Wasley, and Zach (2011) examine contribute to this literature because they help determine whether the relationship between firm-level innovation and operating performance is conditional on the success of a firm’s R&D efforts The conditioning is important because it can sort out more successful from less successful R&D firms The key incremental contribution of Pandit, Wasley, and Zach is the examination of the relationship between the variability of future earnings and patent count and citations that has not been examined in prior studies The relationship between the level of future earnings and patent count and citations has been examined in prior studies (Gu [2005]; Matolcsy & Wyatt [2008]) In the following discussion, I will focus on the empirical models Pandit, Wasley, and Zach (2011) employ and on the reported results Models Pandit, Wasley, and Zach (2011) employ empirical models for explaining the level and variability of future earnings and operating cash flows that are similar to those employed in prior related studies In general, they have expanded the models estimated in related prior studies (such as Kothari, Laguerre, & Leone [2002]; Gu [2005]; Matolcsy & Wyatt [2008]) by including not only the R&D output measures but also the interactions between R&D input and output measures My main comments are on the level of future earnings and cash flow model Pandit, Wasley, and Zach (2011) refer to the Cobb-Douglas production function that suggests a relationship between inputs and outputs In theory, inputs are labor and capital while outputs are economic profits (value added) Capital consists of both tangible and intangible assets Pandit, Wasley, and Zach (2011) appear to proxy tangible capital with lagged capital expenditures and intangible capital with lagged R&D and advertising expenditures However, they not include a labor variable in their profitability model To what extent this omission affects the empirical results is not clear Doogar and McHugh (2009) estimate a Cobb-Douglas production function in which labor is highly significant along with property, plant, and equipment and R&D expenditures If the labor variable highly correlates with the R&D variables of interest, then the results reported by Pandit, Wasley, and Zach (2011) may be significantly biased Another feature of the operating performance model Pandit, Wasley, and Zach (2011) estimate is the inclusion of the lagged net income or cash flow as an independent variable This is a time-series feature that I not think is consistent with the theoretical Cobb-Douglas production function Of course, the lagged net income and cash flow variables are highly significant in the results Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 DISCUSSION—THE EFFECT OF RESEARCH AND DEVELOPMENT 147 reported in Table However, the interpretation of the results differs between a Cobb-Douglas model and a time-series model In the Cobb-Douglas model, the factors of production, capital, and labor cause future profitability and their coefficients are the total effects of the factors on profitability In the time-series model, the coefficient of the lagged net income or cash flow indicates the persistence of these values into the future Thus the significance of any other lagged independent variables (e.g., R&D expenditures, capital expenditures) implies that these variables convey information about future operating performance that is incremental to that conveyed by lagged net income or cash flow Therefore, Pandit, Wasley, and Zach (2011) have estimated incremental effects and not total effects of their variables of interest—R&D inputs and outputs—on future operating performance Obviously, these incremental effects not indicate the ‘‘economic value of innovation’’ that the authors want to approximate Results Pandit, Wasley, and Zach (2011) report results indicating that patent CITATION (quality of innovation) positively associates with future net income and cash flow (Table 3) This is consistent with their expectations and with results reported by Gu (2005) and Matolcsy and Wyatt (2008) However, the results for COUNT are not consistent with the authors’ expectations because in Model of both Panels A and B of Table its coefficient is insignificant In addition, the results appear to be counterintuitive, because in Table 3, Models and COUNT negatively associates with future net income and cash flow The authors not attempt to explain this result, but it needs an explanation in a future study One possibility is that it is the result of high colinearity given that COUNT and CITATION have a correlation of 99 percent (Table 2) The results for R&D expenditures in Table are also mixed R&D is not significant in predicting future net income (Panel A) A potential explanation for this is that past net income already conveys information about past R&D expenditures that is sufficient for the information conveyed by current R&D expenditures Indeed, the results in Table 3, Panel A show that 56 percent of past net income persists into the future Perhaps, such a level of persistence reflects, on average, the benefits of R&D spending However, it is more difficult to understand the significantly negative coefficients of R&D expenditures in Panel B The coefficients imply that $1.00 of current investment in R&D reduces the average future cash flow by $0.15 This is another issue that needs investigation in a future study One possibility is an incorrectly specified model whose estimation leads to counterintuitive results The interaction of COUNT and CITATION yields significantly positive coefficients in both Panels A and B of Table 3, consistent with expectations However, the interaction of R&D expenditures with COUNT and CITATION yields insignificant effects in both panels I find this a surprising result because for Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 148 JOURNAL OF ACCOUNTING, AUDITING & FINANCE CITATION, at least, intuition suggests that firms with high R&D expenditures (high effort) and high patent citations (high innovation quality) should enjoy higher benefits Again, the weak results in this case may be due to either the presence of lagged net income and cash flow in the models or the estimation of incorrectly specified models or both of these reasons Although statistical significance of the R&D variables is present in Table 3, economic significance is another important feature that should be considered For this, I will use the significant estimated coefficients of Model and the means of the related variables (from Table 1) to calculate the mean effect of R&D activity on future profitability (economic value of innovation) in this sample Thus, from Panel A of Table 3, we obtain the following: À0.015*17.05 þ 0.008*0.546 þ 0.003*(17.05*0.546) = À0.224 From Panel B of Table 3, we obtain the following: À0.015*17.05 þ 0.013*0.546 þ 0.031*(17.05*0.546) À 0.165*0.067 = 0.029 Obviously, there is a contradiction between these two results one being large negative (À22.4%) and the other small positive (2.9%) This once again points to the need of a theoretically well-specified model whose estimation should yield more accurate and economically intuitive results My final comment on the results relates to the reported adjusted R2 in both Tables and The R2 increment moving from the partial Model to the full Model is at the third decimal point Evidently, the incremental power of CITATION and the interaction effects is surprisingly very small Since prediction models are estimated, this very low in-sample prediction performance most likely will yield very low out-of-sample performance as well Future Research The paper raises several issues that could be explored in future research The most important issue is empirical model specification I think this issue is critical for the earnings and cash flow prediction literature in general Different prediction models are used by different researchers For example, within the R&D-related literature, Gu (2005) regresses earnings at tþ1 on earnings at t and the change in patent citations from t-1 to t, while Matolcsy and Wyatt (2008) regress earnings at tþ1 on accruals, cash flows, and technology variables at t I think the use of theory is crucial in developing a basic prediction model One avenue is the development of a theoretically well-specified operating performance prediction model based on the CobbDouglas production function For example, an extension of the model employed by Doogar and McHugh (2009) using patent variables may yield reasonable results Another possibility for the development of a prediction model is suggested by equity valuation theory Specifically, Ohlson (1995) develops the following model: " VtT T À1  ðq À 1Þ E T X s¼1 ~ tþs þ X T X # ðq TÀs À 1Þ ~dtþs s¼1 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 ð1Þ DISCUSSION—THE EFFECT OF RESEARCH AND DEVELOPMENT 149 In eq (1), equity value Vt equals aggregated cum-dividend future earnings capitalized at the rate of r-1 where r is plus the discount rate This model is derived from the dividend discount model by assuming clean surplus accounting and it suggests the variable to be predicted: aggregated cum-dividend future earnings (the term in the square bracket) Then, a potential prediction model can be as follows: Aggregated future earnings = f (aggregate past R&D expenditures, aggregate past capital expenditures, aggregate past advertising expenditures, patents, ) In this model, past flows in the form of expenditures predict future flows in the form of earnings An advantage of the model is the aggregation of earnings over a forecast horizon, T Such aggregation captures the benefits of past expenditures over multiple future periods and makes the dependent variable smoother as transitory earnings components tend to cancel out over a long horizon Of course, a disadvantage of estimating such a model is the requirement of a long series of data for each sample firm Another issue that needs investigation in future research is the patent data used For example, COUNT may need to enter into the model in a particular way given its discrete nature and the use of self-citations instead of total citations to patents may be more informative (Hall, Jaffe, & Trajtenberg [2001]) More issues related to data are raised by Seethamraju (2005) in his discussion of Gu (2005) A final suggestion is the use of out-of-sample analyses after in-sample analyses Because in this line of research prediction models are developed and estimated, the quality of the in-sample analysis can be evaluated only by an out-of-sample prediction analysis REFERENCES Doogar, R., and M McHugh 2009 ‘‘Measuring Knowledge Capital.’’ Working paper, University of Illinois Gu, F 2005 ‘‘Innovation, Future Earnings, and Market Efficiency.’’ Journal of Accounting, Auditing and Finance 20: 385À418 Hall, B H., A B Jaffe, and M Trajtenberg 2001 ‘‘The NBER Patent Citations Data File: Lessons, Insights and Methodological Tools.’’ NBER Working Paper 8498, National Bureau of Economic Research, Cambridge, MA Kothari, S P., T E Laguerre, and A J Leone 2002 ‘‘Capitalization versus Expensing: Evidence on the Uncertainty of Future Earnings from Capital Expenditure versus R&D Outlays.’’ Review of Accounting Studies 7: 355À382 Matolcsy, Z P., and A Wyatt 2008 The Association between Technological Conditions and the Market Value of Equity.’’ The Accounting Review 83: 479À518 Ohlson, J A 1995 ‘‘Earnings, Book Values, and Dividends in Equity Valuation.’’ Contemporary Accounting Research 11 (2): 661À687 Pandit, S., C E Wasley, and T Zach 2011 ‘‘The Effect of Research and Development (R&D) Inputs and Outputs on the Relation between the Uncertainty of Future Operating Performance and R&D Expenditures.’’ Journal of Auditing, Accounting and Finance 26 (Winter): 121–144 Seethamraju, C 2005 ‘‘Discussion-Innovation, Future Earnings, and Market Efficiency.’’ Journal of Accounting, Auditing and Finance 20: 419À422 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2012 [...]... 8.8% 10 .5% 4.9 11 .0% À0 .18 10 .2% 1. 08 4 .1% 1. 01 5.6% 6.8% 1. 19 10 .1% 408 2 .11 16 .3% À7 .1% 9.2% 10 .2% 5.3 10 .9% À0 .15 10 .0% 1. 09 4.2% 1. 01 5.5% 6.8% 1. 15 9.7% 408 0.82 14 .9% 13 .6% 1. 3% 19 .9% 0.4 11 .0% À0.80 29.5% 0.37 3.0% 0.99 7.2% 8.0% 1. 79 28.4% 408 2 .10 16 .2% À9.2% 7.0% 9.2% 4.4 10 .8% À0.32 11 .4% 0.95 4.3% 0.99 5.4% 6.5% 1. 31 11. 4% 408 2.58 16 .1% À8.2% 7.9% 10 .3% 4.5 11 .2% À0.28 13 .6% 0.82 4.3%... December 2, 2 012 12 .7% 12 .7% 0.0% 6.2% 12 .7% 15 .4% 0.45 24.8% 37.5% 22 .1% 1. 44 34.3% 0.25 21. 8% 1. 58 37.5% 12 .7% 24.8% 2.54 19 .6% À9.9% 9.7% 9.5% 6.0 13 .4% À0.29 8.3% 1. 61 7.5% 38.5% 59.4% 28.7% 1. 87 52.8% 0.52 27.5% 1. 92 59.4% 12 .7% 46.7% 5 .15 18 .5% À9.4% 9 .1% 5.5% 9.6 10 .3% À0.09 5.3% 1. 94 4.4% 48.4% 74.7% 31. 7% 2 .18 73.0% 0 .13 31. 6% 2. 31 74.7% 12 .7% 62.0% 5.34 20.2% À8.6% 11 .6% 6.4% 10 .6 13 .7% À0 .17 5.9%... 2 012 VW Market 408 PBCs 0.24 0.06 408 ABCs 12 .7% 12 .7% 0.0% 6.2% 12 .7% 15 .4% 0.45 24.8% 18 .6% 17 .1% 0.76 6.4% 0.96 8.6% 0.75 18 .6% 12 .7% 5.9% 7.9% 14 .4% 18 .3% 0.47 0 .1% 1. 12 5.9% 0.02 14 .4% 12 .7% 1. 7% 24.9% 18 .7% 17 .2% 0.76 6.5% 0.96 8.6% 0.76 18 .7% 12 .7% 6.0% 0.05 À0.06 408 PBCs þ ABCs 23.6% 18 .5% 17 .0% 0.76 6.4% 0.96 8.4% 0.76 18 .5% 12 .7% 5.8% 408 PBCs 5.6% 13 .4% 16 .2% 0.48 0.3% 1. 03 2.5% 0 .12 13 .4%... 1. 66 À2.58 408 35.0% 58.0% 30.5% 1. 71 50.6% 0.59 29.2% 1. 74 58.0% 12 .7% 45.3% 5.02 18 .5% À9.7% 8.8% 4.6% 11 .2 9.8% À0.08 5.2% 1. 88 4.3% ABCs 1. 97 À2 .13 408 43.6% 75.5% 37.0% 1. 89 74.6% 0.07 37.0% 2.02 75.5% 12 .7% 62.8% 5.49 19 .9% À8.6% 11 .3% 5.9% 11 .2 13 .2% À0 .16 6.2% 2 .13 6.3% PBCs þ ABCs PPP Out -of- Sample Monthly over January 19 75–December 2008 23.5% 42 .1% 26.7% 1. 36 40.5% 0 .12 26.7% 1. 52 42 .1% 12 .7%... 2, 2 012 VW Market 408 PBCs 0 .19 À0.85 408 À0.05 À0.88 408 Certainty Equivalent r Mean r s (r) Sharpe Ratio a b s (e) Information Ratio Optimized Mean r ¼ A Mkt r ¼ B Managed rlev ¼ A – B 12 .0% 12 .0% 0.0% 5.9% 12 .0% 15 .2% 0.42 9.2% 17 .4% 17 .6% 0.67 6.5% 0. 91 10.9% 0.60 17 .4% 12 .0% 5.4% 17 .6% 31. 0% 22.0% 1. 15 20.9% 0.84 18 .0% 1. 16 31. 0% 12 .0% 19 .0% 18 .5% 32.6% 22.9% 1. 18 22.0% 0.88 18 .6% 1. 19 32.6% 12 .0%... 1. 25 9.9% 408 PBCs 1. 11 10.8% 5.9% 1. 20 10 .2% 408 ABCs 1. 06 10 .5% 7.0% 1. 23 10 .3% 408 PBCs þ ABCs 0.96 10 .0% 6.0% 1. 25 10 .9% 408 PBCs 1. 11 10.7% 5.5% 1. 19 10 .2% 408 ABCs 1. 07 10 .9% 6.9% 1. 23 10 .4% 408 PBCs þ ABCs PPP Out -of- Sample Monthly over January 19 75–December 2008 Note: This table reports the results of estimating the coefficients of the linear portfolio weight function specified in eq (3)... 1. 19 408 PBCs þ ABCs PPP Out -of- Sample Monthly over January 19 75–December 2008 8.8% 14 .4% 20.2% 0.43 4.3% 0.84 15 .7% 0.27 14 .4% 12 .0% 2.4% PBCs 408 PBCs PPP Out -of- Sample Monthly over January 19 75–December 2008 Downloaded from jaf.sagepub.com at Taylor's University on December 2, 2 012 0.77 15 .0% À9.2% 5.7% 23.9% 1. 4 11 .4% À0.47 22.8% 0.50 2.8% 1. 01 8.3% 8.5% 1. 48 20.8% 408 2.03 16 .3% À7.5% 8.8% 10 .5%... of Financial Studies 17 (Spring): 63–98 Alexander, G J 19 93 ‘‘Short Selling and Efficient Sets.’’ Journal of Finance 48 (September): 14 97 15 06 Asness, C., R Krail, and J Liew 20 01 ‘‘Do Hedge Funds Hedge?’’ Journal of Portfolio Management (Fall): 6 19 Ball, R J., and P Brown 19 68 ‘‘An Empirical Evaluation of Accounting Income Numbers.’’ Journal of Accounting Research 6 (Autumn): 15 9 17 8 Banz, R 19 81. .. 19 75–December 2008 0 .16 3.0 À0.5 À2.6 0.45 À0.20 1. 73 0.73 À3.0 [3.6] 22.8 [4.5] 11 .4 [4 .1] PBCs 1. 71 16 .7 [7.3] 29.7 [10 .5] À33.0 [8.8] 0.27 3.3 1. 0 À4.7 0.48 ABCs À4.9 [4.2] 21. 4 [6.5] 2.4 [6.0] 19 .8 [8.2] 35.4 [12 .0] À22.9 [8.9] 0.29 3 .1 1. 2 À5 .1 0.48 À0.44 1. 65 0. 61 1. 65 PBCs þ ABCs PPP Out -of- Sample Monthly over January 19 75–December 2008 Brandt, Santa-Clara, and Valkanov’s (2009) Linear Parametric... Reinganum, M R 19 81 ‘‘Misspecification of Capital Asset Pricing: Empirical Anomalies Based on Earnings’ Yields and Market Values.’’ Journal of Financial Economics 9 (March): 19 –46 Rosenberg, B., K Reid, and R Lanstein 19 85 ‘‘Persuasive Evidence of Market Inefficiency.’’ Journal of Portfolio Management 11 (Fall): 9 17 Sharpe, W F 19 94 ‘‘The Sharpe Ratio.’’ Journal of Portfolio Management 21 (Fall): 49–58