E/B/R ECONOMIC AND BUSINESS REVIEW CONTENTS 313 375 423 Aleš Pustovrh Marko Jaklič Intellectual Structure of the Open Innovation Field: State of the Art and a Critical Literature Review 347 Nathan H. Jeppson Matthew C. Geiszler David F. Salerno Firm Credit Ratings and Financial Analyst Forecast Performance Valentina Božič Ljubica Kneževic Cvelbar What Really Defines the Performance in Hotel Industry? Managers' Perspective Using Delphi Method 395 Katja Babič Matej Černe Miha Skerlavaj Pengcheng Zhang The Interplay among Prosocial Motivation, Cultural Tightness, and Uncertainty Avoidance in Predicting Knowledge Hiding Hasna Chaibi Research and Development Expenses under IFRS Mandatory Implementation: A Value Relevance Approach INTELLECTUAL STRUCTURE OF THE OPEN INNOVATION FIELD: STATE OF THE ART AND A CRITICAL LITERATURE REVIEW Ales Pustovrh1 Received: October 3, 2016 Marko Jaklic2 Accepted: April 25, 2018 ABSTRACT: This paper presents an overview of open innovation. It positions the concept into a wider framework of scholarly research of innovation, discusses its historical development and its positioning within the wider area of innovation research. Using different types of bibliometric analysis, we estimate the impact of open innovation and continue with their contributions to the theory of innovation. While not a true paradigm shift as it used to claim, open innovation is a clearly defined innovation concept that brings important contributions to the theory of innovation, helps answering some of the key questions that were recognized by innovation scholars. Keywords: open innovation, innovation concepts, open innovation practices, open innovation policies, user innovation, bibliometric analysis JEL: O31, I24 DOI: 10.15458/85451.68 1. INTRODUCTION The nature of innovation has changed recently. Innovation activities have become globalized and open in a way that was unimaginable even 20 years ago (Wooldridge, 2010). Companies now innovate in an environment in which competition is global, knowledge is spread more widely, R&D investments are increasing and in which product life cycles are shortening (Koen De Backer, Cervantes, Van De Velde, & Martinez, 2008). Companies can no longer succeed by developing the next innovative product in their internal laboratories or by outsourcing manufacturing activities to low-cost countries (Herrigel, 2010). Countries implement competing innovation policies in order to become more attractive as potential innovation hubs. These changes have brought new insight into innovation research. Several theoretical concepts have emerged, but the most interest has recently been devoted to a new innovation concept of open innovation, introduced by Henry Chesbrough's 2003 book (Chesbrough, 2003). The open innovation concept presumes that companies use external ideas besides 1 Corresponding author, University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: ales.pustovrh@ef.uni-lj.si 2 University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: marko.jaklic@ef.uni-lj.si those generated inside the boundaries of the company. They also seek internal and external ways to the market for them. Research and development represent an open system (H W Chesbrough & Crowther, 2006)3. The underlying reason for the development of open innovation and other theoretical innovation concepts was the changing nature of innovation practices. During the first decades of the 20th century, industrial enterprises in the US cooperated and sourced R&D services from dedicated external R&D labs in a way that is very familiar to the current practitioners of open innovation (E. K. R. E. Huizingh, 2011), (Mowery, 1983). The best known example is Edison's The Invention Factory at Menlo Park. Cooperation between companies was common at the time and critical to the survival of an industrial structure dominated by small firms (Hollingsworth, Campbell, & Lindberg, 1991). Still, there was a large gap in theoretical understanding of innovation that was being observed in innovation practices. Open innovation concept has been targeting this lack of understanding observed in existing innovation practices. Chesbrough claims that open innovation represents a paradigm shift. It emphasises cooperation and sharing of ideas between companies regardless of the boundaries between companies or states. Companies buy or license processes and innovations from other companies and at the same time push their innovations to the market through licensing, joint ventures or spin-offs (Chesbrough, 2003). This challenges the 'closed' innovation model that sees innovation as the result of work of the large internal laboratories that only large, usually multinational companies can afford. While no one disputes that the open innovation concept has attracted a lot of attention both in practice and academia (E. K. R. E. Huizingh, 2011), there are authors that claim it is not a clear concept and that it comes in many forms, which makes the concept rich but hinders generalization. Others dispute the paradigm shift that open innovation claims to present. They predict that the term will fade away in a decade (E. Huizingh, Conn, & Torkkeli, 2011), merging into the 'standard' definition of innovation. Others have suggested that the term itself could be acting as a communication barrier - hindering growth in research and understanding, thus representing constraint to future research (Groen & Linton, 2010). Based on these insights, our analysis on the literature review aims to contribute to filling the gaps in understanding innovation recognized by scholars (Fagerberger, 2005) and to answering three research questions in particular: 1. Is it really a new paradigm in understanding innovation (Chesbrough, 2003)? 2. Is open innovation a new innovation concept or just the continuation of the innovation research and not distinct from other existing innovation concepts (E. Huizingh et al., 2011)? 3 In the paper, we continue to use Chesbrough's definition of open innovation which he defines as: "the use of purposive inflows and outflows of knowledge to accelerate internal innovation and to expand the markets for external use of innovation, respectively" (Henry William Chesbrough, Vanhaverbeke, & West, 2006, p. 1). 3. Is open innovation even important for theory and practice of innovation - or is the term a communication barrier hindering growth in innovation research and understanding of innovation (Groen & Linton, 2010)? We will try to provide answers to these three research questions using the bibliometric methods and critical literature review. 2. DEVELOPMENTOFTH E OPEN INNOVATION CONCEP TIN INNOVATIONRESEARCH The beginning of the open innovation concept is clear - , it was introduced in Henry Chesbrough's book in 2003 (Chesbrough, 2003). It received significant interest from scholars who soon followed with a growing number of publications. Figure 1: Number of papers on open innovation and its share among innovation papers Source:ThompsonReuters Web of Imowledge wie The initial studiewoR og eninnowgtiwnfocuse d on earlyadopterw and gocd practiee wxamplee.wRich tendtobe SThenssWul Theywereusualtycasectudiec anV descriptixe in natutwe.g. (Hudton & Sakkab, 2006), (Chiaroni, Chietn, W Fratiini, bOOt) on (Chrisiexeen, 01rcen,&Kjar,2005i.Mpctcas rstudiessiso fnciretonp aeticei^!tr^nai^;^^sie^,mhtthii^^n nigh)2ech(CdesFrodgh, 2003). Theee initial stoUies were followed b)n expanding the seepe tpi oiher indusiries (H W Chesbrough & Crowther, 2006). Some authors soon discovered that 'few corporations have institutionalised open innovation practices in ways that have enabled substantial e;eowthorindustlxleglgenshiR' (RuM-Latre, Mullee, w inxes,20ra). Mpre cast studies followed that attempted to compare different open innovation practices to determine their context dependency e.g.(Sarkar & Costa, 2008), (Vanhaverbeke, Ine, & De Zutter, 2012). They expanded the scope of activities connected to the concept (Gr0tnes, 2009). At the same time, the first empirical studies were implemented. They initially used existing data sources like the European CIS survey (Ebersberger, Herstad, Iversen, Kirner, & Som, 2011), (Mention, 2011) or global indicators that were not designed to measure open innovation (K. De Backer, López-Bassols, & Martinez, 2008). They later included specific quantitative studies, but often focused on certain industries (Harison & Koski, 2010), countries (Lazzarotti, Manzini, & Pellegrini, 2010) or institutions (Spithoven, Clarysse, & Knockaert, 2010). Some quantitative studies focused on small and medium sized companies and discovered that open innovation is a logical step for them. Consequently, they are collaborating with external partners more frequently than large companies (van de Vrande, de Jong, Vanhaverbeke, & de Rochemont, 2009) Others discovered that open innovation is not always the best option (Praest Knudsen & B0tker Mortensen, 2011). A 'closed' innovation system can also be more suitable for some companies or even industries, as there are clear differences in open innovation among companies and industries. Some authors even believe that 'closed' innovation systems could return and see evidence of that emerging with the development of the Internet (Anderson & Wolff, 2010). In recent years, open innovation research has seen systematic appraisals ofthe contributions of the open innovation in the form of several literature reviews and summaries, as well as identifications of areas for future research (West & Bogers, 2017). As the table X shows, the body of research on open innovation is still increasing and that means that it is spreading to new areas and targets new research questions. Interestingly, one of the opportunities identified by scholars is also a better connection to prior theoretical research, including topics such as absorptive capacity, user innovation, resources, dynamic capabilities, business models, and the definition of the firm (West & Bogers, 2017). They clearly recognized the opportunities of expanding the scope of open innovation beyond organisational-level research to multiple levels of analysis (Bogers et al., 2017). There is some evidence that open innovation is limited to certain research areas - for example R&D management. Analysing the interest in open innovation, the search of Thomson Reuters Web of knowledge for "open+innovation" after 2003 (when the concept was established) and limiting the results to management, business and economics categories resulted in 1.554 documents in our document set. We have limited our research to the business, management and economics categories because they are by far the most numerous ones. Other categories are numerous but very limited, as the graph shows. Figure 3: Open innovation topic among the Web of knowledge categories Source:ThompsonReutersWeb ofknowledge2018 Our analysis also shows thatopen innovatio n remainsmostwidelyuse O inth eman^a^^i^nie nt of l^ikD^.Thi^^iscleai^ly seen fioun i^hen^(3ojsu]^i^als;p^e^biisU^nac^t^(^jii mnsivationrese aidi withfouront otfioa in oto ucnais tiistpublish ape n^^a^i^oae^ioc_iese r^eaOfDC usioaonR&D management. Table 1: Top journals that publish open innovation research Source Titles Published articles R D MANAGEMENT 67 RESEARNA TECHNOLOGY MANAGEMENT 59 TECHNOVETIAN 5 3 RESEAGCHGELIEG 51 INTERVALGONTL JOUENALOFTECHNO LOGEMLNAGEMENT 5 0 TECHNNLGTY ANALYSIG STRATEGICMAOTGEMNNT 4 6 TECHNOLGEIOAL ^O^^C^iRO'^I^^ AND GOCIAL LVLVGE 4 2 JOURNAL TIP I^^OI^I^COG^ITVE^^T^IONGIENAGEMENr 3 7 CREATIVITGAND EIOTOTG^ON MGNTGEMENO 3 2 INTERNATIONAL JOURNALOFINNOVATIONMANAGEMENT 32 Source: Thompson Reuters Web of knowledge 2018 'Hiefoeus onbnsinessanomoslencmero assets ouonuTinoovttionisoaunst:ovl sndnot uotngino duo01 ^r^aah^^ia^ooi^sieae^u^^ni^lsc lIutmrss ^^a^^rc;1! topks. Eventine tat:euacy s^t nconomits is noe wnll cucuaaehsd- and the interest of open innovation scholars in economics seems even to be waning in recent years. Figure 4: Categories chosen by the open innovation articles Source:ThompsonReutersWebof knowledge 2018 Whileouranalysisfocusesonthe articles publishedafter 2003, some articlesalsoresults froae eur se^^ei^lf shai ett^^^^^fed before OiOOt^. Oa^t^i^irastingty, m the saose citcd leteratueo ^e^aws o^cas^ d.ipcor.od amcnp finann.al analyo1ts' Wink PEIs depsndonW vrnicbk ir io1 comporad ee. are actual meieure os esmin..0, dkpresion ean bnmteepretei^s o pr 2xeiebc sl.fferenceoinojtiinion omoTofinantialunal:ostf fore oecurity (Diather rt =L,U0P2t cndl Iharafora,oncarteintyineor]oinaryore8ssSs.Prigr studies 8lsofinP ihsi widediapeoeioya8tte^rsinforrcast values indicate complexity in making 3)0^^0 fooefarts(DcEU T Reel^ r0 02. Marri21^. DeFigueu'uclo^O Id Tan, & Welkar; 200 b). 3.2. Independent Variables The credit rating variable (CR) is the S&P Domestic Long-term issuer credit rating from Compustat (mnemonic: SPDRM). It is defined as the Standard & Poor's Issuer Credit Rating which represents the "current opinion of an issuer's overall creditworthiness, apart from its ability to repay individual obligations. This opinion focuses on the obligor's capacity and willingness to meet its long-term financial commitments (those with maturities of more than one year) as they come due." CR ranges from one to 22, with one representing a D rating and 22 representing a AAA rating. Therefore, higher values of CR denote that a firm has received a higher credit rating from S&P. 3.3. Control Variables We include the controls FOLLOW, LOSS, AEARN, HORIZ, and VOL from prior literature (Duru & Reeb, 2002; Mauri, Lin, & DeFigueiredo, 2013), all of which except for FOLLOW control for complexity and uncertainty in the forecasting task. We also include controls for the quality of reported earnings (EQ) and for the smoothness of reported earnings (SMOOTH), both of which could also introduce uncertainty into the forecasting environment. Finally, following Duru and Reeb (2002) and Bae, Tan and Welker (2008) we also include controls for industry and fiscal year. FOLLOW is measured as the total number of financial analysts following the firm for a given period. The typical expectation is that larger financial analyst following is associated with lower forecast optimism and greater accuracy (Drake & Myers, 2011; Duru & Reeb, 2002; Mauri, Lin, & DeFigueiredo, 2013). Furthermore, Lang and Lundholm (1996) indicate that larger following is associated with decreases in overestimates because financial analysts prefer to follow firms with high quality earnings reports. However, Duru and Reeb (2002) find greater following to be positively and marginally associated with forecast accuracy and not a significant factor for overestimates. In light of the above studies, if significant, we expect FOLLOW to be negatively associated with dispersion and positively associated with accuracy. LOSS (an indicator variable coded 1 if the firm reported a net loss in the period, otherwise zero) is included because managers of firms that report losses may tend to understate net income in order to enhance net income in subsequent periods; a practice known in the literature as taking a "big bath" (Elliott & Hanna, 1996; Moehrle, 2002). Firms that engage in such practices understate net income which causes forecasts to be higher than reported income. Additionally, Brown (2001) finds that financial analysts issue more optimistic forecasts in periods of losses, and Hwang, Jan, & Basu (1996) find that financial analysts are less accurate while forecasting losses than when predicting positive net income. However, Duru and Reeb (2002) find that LOSS is not significant for accuracy. We expect LOSS to be positively associated with dispersion and negatively associated with accuracy. The AEARN variable is equal to the absolute value of the change in earnings per share from the previous year divided by the stock price at the beginning of the year. Duru and Reeb (2002) find that larger absolute changes in earnings per share are negatively associated with accuracy. However, because large changes in earnings could present uncertainty, we do not predict a direction. Following prior literature the forecast horizon control (HOJRI20 F^p^At^s-^^Shp i^cn^3^iSFiLLca^s^ETd^i'R^^^s;R,Tt\Tee^sc thefoT^RETH iLNuedateandStiT actua1 cp^rs^inEnai^np^un^cFETFL^t:d^Ie. The lippci^1:u^FE shows thatLsn^r foreLaPI hc^i^p^coR^c areassociatedwIcrt less actNFDte(BrAwn,20 01;Clement&Tse , 2003iJapob.Lys, &NFAIr, ^icSI9) anpptiF^i^eCiRpEo.i^]pe^iTLtL3^cJSpcec^Is1A^uru&IE(^Tb,^(^(^2; Monsi, Maowe((, & MiUec, apgS). Whexpecp louncp horizonrto be aesNEicteOwiFAmorc BisE;erEiunonA Irss accocapy Solatia sorniwgsare lmowntoOT dssnpipPeawiCS lees LTCUFate eRtnings fosFcasisiDuru & RFeb,2R02; Lim, MO1; Ksi] rne romddny dssed on ilse tlB/ElS S21n coNeoe]End cEASl hepreeents ii:niii_Lal;or variDn0bL fci tiete firms fidtai brao] bates)) hCopotSeesee I1 anil 3o we rrtimahe eiestat^oro (5)° HnI1, oltlo ttiieL ^ 0 anticipates d positive 0Le0tila5]icIlt bciwoen oarnlndr Ool•acast octucacyaddfirma•edtt satingsas wollos anegctive aesoaialion bofwoen enrmngrfoeecaeb dt^r^^i^^l^n snd firmc8ecLitratlngSt ^yDo^lr^^i ^O oda awn1 es li Nel5 ative oam o a1 etio n de2WDon e armngsqo aho1 an i1 .ooecast aoo urac3 cswoUos o pIlLitihe arroolation betwoen Darnings oiuahty sad earnings fbaecasL dtspesr1on.EqdaUisn ifl te eslimat csL as 0'o)l_olosl Cdeseesttia +2 Lmwt o Liuste dLaodnnteito dL(nooos dL^ruheeh,- + L8H0RCZs tt L7hOns d ^sdh^Cifj +L9YUARv (5) wdiier LA ss thd ^adeS taLius; vdriobte and oangem Oimom onu te 22 lono ee]3:3asei^tlo^gdD rating md wrth ^reprosentimga .AA.A. r at1!^3;] ^Q repots ^nte t0o loq sW the standard deviation of the firm's level of earnings quality over the prior five consecutive years, and aa atiseo variobks cre ckfino d ms in vquotion io1) . FivalSy, to testHbpo5hescL4,wnicODnticipo3eethatat hiuVoL ievaasoUsredlN oatinge, the negitive ess ociot:iNnbb3we e^o^eninge nuohty uad finonclcl analystforecar3 ooodscayor well we lho dbLitiveasoociutidn betweeneoonlngs ouallty and fi^^nEial^nc^is^^i forecast di^ersmn^ seiluced, ■see estimate a qurtiool (6l. We tnolude tile mteruetion o f credl) ratings wi5tieamlno>s quahtyiCrieeiQ) 1nourbasfcmddrlsaseoNowe: Ce^estest = a o LiCWf o LiUgf dLjCeougf oLcFonnodf dLinooOt dL^ruAWht g L7HOwe6f o Lshont o L^ehco L;<>erwswf (6) 3.5. DetaS ample We collect a sample of company level reported data, including credit ratings, for the period 2006 to 2015 from the Compustat Annual Database. We combine individual financial analyst forecast data from the Institutional Brokers' Estimate System (I/B/E/S) for the same period. After merging data from these sources, we omit any firm-year observations with insufficient data, any firm-year observations with a financial analyst following of less than two, with a horizon of more than one year, and then limit the sample to the most recent forecast for each financial analyst in each firm year. The resulting full sample contains 100,137 firm year observations. This sample is used in our testing of Hypothesis 1. We then reduce the sample to only observations which contain a Standard and Poor's credit rating for each firm-year. This reduces the sample size to 88,652 observations. This reduced sample is utilized to test the remaining Hypotheses 2 - 4. Table 1 Descriptive Statistics (n=100,137) Variable Mean Median Standard deviation 25 th Percentile 75th Percentile AFEP -8.9569 -0.1772 1329.4 -0.5613 -0.0573 DISP 0.5133 0.0048 88.730 0.0021 0.0119 CR 8.1187 10.000 7.1398 0.0000 14.000 EQ 0.0333 0.0245 0.0324 0.0160 0.0396 FOLLOW 19.891 19.000 11.156 11.000 27.000 LOSS 0.1649 0.0000 0.3711 0.0000 0.0000 AEARN 1.2643 0.0119 223.74 0.0056 0.0277 HORIZ 103.86 97.000 81.971 43.000 116.00 VOL 0.0626 0.0318 0.1485 0.0160 0.0661 AFEP is the absolute forecast error multiplied by -100. DISP is forecast dispersion, measure using the standard deviation of analysts' earnings forecasts deflated by the stock price on the forecast date. CR is the Standard & Poor's Issue Credit Rating which is the opinion of an issuer's overall creditworthiness with codes ranging from 1 representing a D rating and 22 representing a AAA rating lagged by one period. EQ is the earnings quality of the firm calculated as the lag of the standard deviation of the residuals from firm-specific regressions of changes in working capital on past, present, and future operating cash flows over the prior five consecutive years. FOLLOW is the number of analysts following the firm per I/B/E/S. LOSS is an indicator variable coded as 1 for firm-year observations with negative earnings, and 0 otherwise. AEARN is the change in earnings, measured as the absolute value of the difference between the current year's earnings per share and the previous year's earnings per share. HORIZ is the forecast horizon, expressed as the number of days between the forecast and the end of the fiscal year. VOL is earnings volatility, measure as the standard deviation of earnings per share for the previous 5-year period. Table 1 provides the descriptive statistics for the test variables of our models. The mean and median values of -8.9569 and -0.1772 for AFEP are negative by design and indicate the difference between a financial analyst's earnings forecast and the actual earnings. The median value for CR is 10, the equivalent of a BB- rating which Standard and Poor's defines as indicative that the obligor faces major ongoing uncertainties. The average forecast horizon for the sample is only 103.86 days or more than three months. The mean and median values for FOLLOW are 19.891 and 19 and the 25th and 75th percentiles are 11 and 27 indicating that the majority of the firms in the sample have more than a dozen financial analysts providing estimates for the firms for any given firm-year. The variable EQ has mean and median values of 0.0333 and 0.0245 as measured using the lag of changes in working capital on past, present and future operating cash flows over the prior five consecutive years. While the period for our sample covers the years from 2006 to 2015, this variable draws on data that extends back to the year 2001 due to its use of information from prior periods to construct firm-specific regressions regarding the mapping of cash flows to accruals. The mean value for LOSS is 0.1649 indicating that only a small portion of the sample includes firms with negative annual earnings. Table 2 Pearson/Spearman Correlations (n=100,137) AFEP DISP CR EQ FOLLOW LOSS AEARN HORIZ VOL AFEP 1 -0.62357 0.11709 -0.07246 0.14784 -0.32649 -0.43561 -0.15724 -0.24791 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 DISP -0.63008 1 -0.15145 0.09655 -0.16367 0.45636 0.54070 -0.01567 0.37320 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 CR 0.19711 -0.24303 1 -0.16883 0.36279 -0.19109 -0.14302 -0.09893 -0.25994 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 EQ -0.12222 0.15802 -0.20313 1 -0.15136 0.07581 0.12569 0.02829 0.22144 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 FOLLOW 0.24023 -0.23126 0.41250 -0.19041 1 -0.13684 -0.17202 -0.08250 -0.12465 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 LOSS -0.31926 0.44566 -0.21774 0.07663 -0.14276 1 0.31492 0.03120 0.38453 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 AEARN -0.39958 0.59119 -0.18572 0.16353 -0.18002 0.29278 1 0.00675 0.35388 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.0326 <.0001 HORIZ -0.16101 -0.02395 -0.15771 0.04341 -0.14183 0.04475 0.00025 1 0.02866 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.9376 <.0001 VOL -0.34879 0.53286 -0.29018 0.27947 -0.11008 0.41424 0.43373 0.00237 1 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.4539 Pearson correlations are shown above the diagonal. Spearman are shown below. Note that the correlation value is shown with the associated p-value immediately below. All variables are defined in Table 1. Table 2 provides the Pearson and Spearman correlations. The Pearson correlations are reported above the diagonal and the Spearman correlations are reported below the diagonal. Most of the correlations between the independent variables exhibit relatively small correlations. The test variables, AFEP and CR are positively correlated for both the Pearson and Spearman measures. Additionally, DISP and CR are negatively correlated for both the Pearson and Spearman measures. This offers preliminary evidence regarding Hypothesis 2. 4. RESULTS Panel A of Table 3 shows the results of estimating equation (4) with AFEP as the dependent variable, while Panel B provides the results of the same model with DISP as the dependent variable. The coefficient on CREXIST is negative and significant at the one percent level for the test with AFEP as the dependent variable, indicating that the existence of a credit rating is associated with less financial analyst forecast accuracy. Similarly, for the test with DISP as the dependent variable, the coefficient on CREXIST is positive and significant. This indicates that the existence of a credit rating is associated with greater dispersion in forecast patterns. As discussed in Section 3, while a credit rating contains information that may improve the forecast performance of financial analysts, the results in Table 3 indicate less accuracy and more dispersion for firms where a credit rating exists. This could be due to the theory that analysts exert more effort in making their forecasts when there is less available information. Therefore, by rejecting Hypothesis 1, which was stated in the null form, we find a significant association between the existence of a credit rating and analyst forecast performance. Table 3 Rating Availability with AFEP and DISP including EQ Panel A Panel B Dependent variable: AFEP Dependent variable: DISP Independent Variable Coefficient Coefficient Intercept 0.3588 *** 0.0000 *** CREXIST -0.0674 *** 0.0006 *** FOLLOW1 10.440 *** -0.0974 *** LOSS -1.1006 *** 0.0163 *** AEARN -14.110 *** 0.1596 *** HORIZ1 -4.1100 *** 0.0015 ** VOL -1.4896 *** 0.0378 *** IND Included Included YEAR Included Included Adjusted R2 0.2752 0.4196 n 100,137 100,137 *, **, *** indicate statistical significance utilizing heteroscedasticity consistent standard errors at the 10%, 5%, and 1% levels, respectively. 'These coefficients are multiplied by 1,000 for ease of interpretation. CREXIST is an indicator variable coded as 1 if the observation includes a Standard & Poor's Issue Credit Rating, and 0 otherwise. IND is an industry control using the I/B/E/S industry classification. YEAR is an annual control representing the year in which firm j's earnings are reported. All other variables are defined in Table 1. Consistent with prior literature, FOLLOW is significant in both tests indicating that when more financial analysts follow a firm, financial analysts are more accurate and are less dispersed in their earnings forecasts. The coefficient on LOSS is significant for all tests indicating that financial analysts are less accurate and more dispersed for firms that experience a loss in the current period. Prior literature demonstrates that financial analysts in some cases fail to predict failures and as a result, firms with losses are more likely to be associated with negative earnings surprises (Skinner & Sloan, 2001; Kinney, Burgstahler, & Martin, 2002). The AEARN control is highly significant for all tests. The negative sign for tests with AFEP as the dependent variable suggests that larger absolute increases in earnings from the prior period are associated with less accurate forecasts of future earnings. HORIZ is significant in both tests of dispersion, indicating that shorter time periods between forecasts and earnings disclosures leads to more accuracy and less dispersion. Finally, as predicted the VOL control is significant in both tests. VOL represents the volatility in a firms reported earnings, thus a negative result for tests involving forecast accuracy indicate that a lack of smooth earnings from year to year increases the likelihood of forecast inaccuracy and a positive result for tests involving dispersion indicate that a lack of smooth earnings increases forecast dispersion. Panel A of Table 4 shows the results of estimating equation (5). The first two columns of results provide the results of estimating equation (5) with AFEP as the dependent variable, while the final two columns provides the results of the same models with DISP as the dependent variable. Note that, as predicted, the coefficient on CR are positive and significant at the one percent level for the test with AFEP as the dependent variable indicating that higher levels of CR are associated with more financial analyst forecast accuracy. This result supports Hypothesis 2 and indicates that financial analysts produce more accurate earnings forecasts for firms with higher credit ratings. Similarly, for the test with DISP as the dependent variable, the coefficient on CR is negative and significant. This also supports Hypothesis 2 and indicates that financial analysts' earnings forecast estimates are less dispersed for firms with higher credit ratings. Table 4 Rating with AFEP and DISP including EQ Panel A Dependent variable: AFEP Independent Variable Coefficient Coefficient Intercept -0.0629 -0.4636 *** CR 0.0352 *** 0.0674 *** EQ 0.9469 ** 19.4343 *** CR*EQ -1.4310 *** FOLLOW1 10.2900 *** 10.3000 *** LOSS -1.3173 *** -1.3158 *** AEARN -19.0601 *** -19.1496 *** HORIZ1 -3.8300 *** -3.8400 *** VOL -2.2852 *** -2.4680 *** IND Included Included YEAR Included Included Adjusted R2 0.3350 0.3363 n 88,652 88,652 Panel B Dependent variable: DISP Coefficient 0.0069 *** -0.0007 *** -0.0183 *** -0.0173 ** 0.0173 *** 0.2170 *** 0.0013 0.0758 *** Included Included 0.4830 88,652 Coefficient 0.0100 *** -0.0010 *** -0.1617 *** 0.0111 *** -0.0173 ** 0.0173 *** 0.2177 *** 0.0014 * 0.0772 *** Included Included 0.4837 88,652 *, **, *** indicate statistical significance utilizing heteroscedasticity consistent standard errors at the 10%, 5%, and 1% levels, respectively. 'These coefficients are multiplied by 1,000 for ease of interpretation. IND is an industry control using the I/B/E/S industry classification. YEAR is an annual control representing the year in which firm j's earnings are reported. All other variables are defined in Table 1. Also as predicted, the coefficient on EQ is positive and significant for the test with AFEP as the dependent variable indicating that lower earnings quality, as indicated by higher levels of EQ, is associated with more financial analyst forecast accuracy. This result supports Hypothesis 3 and indicates that financial analysts produce more accurate earnings forecasts for firms with lower earnings quality due to their management of earnings to match financial analysts' earnings expectations. Similarly, for tests with DISP as the dependent variable, the coefficient on EQ is negative and significant. This also supports Hypothesis 3 and indicates that financial analysts' earnings forecast estimates are less dispersed for firms with lower earnings quality and therefore more dispersed for firms with better earnings quality. Consistent with prior literature, FOLLOW is significant in all four tests indicating that when more financial analysts follow a firm, financial analysts are more accurate and are less dispersed in their earnings forecasts. The coefficient on LOSS is significant for all tests indicating that financial analysts are less accurate and more dispersed for firms that experience a loss in the current period. Prior literature demonstrates that financial analysts in some cases fail to predict failures and as a result, firms with losses are more likely to be associated with negative earnings surprises (Skinner & Sloan, 2001; Kinney, Burgstahler, & Martin, 2002). The AEARN control is highly significant for all tests. The negative sign for tests with AFEP as the dependent variable suggests that larger absolute increases in earnings from the prior period are associated with less accurate forecasts of future earnings. Recall that AEARN is the absolute value of the change in earnings per share from the previous year scaled by price. This result is likely due to financial analysts' uncertainty when faced with large changes in earnings from one period to the next. HORIZ is not significant in our tests of dispersion, but consistent with prior literature is significant for tests involving accuracy. Finally, as predicted the VOL control is significant in all tests. VOL represents the volatility in a firms reported earnings, thus a negative result for tests involving forecast accuracy indicate that a lack of smooth earnings from year to year increases the likelihood of forecast inaccuracy and a positive result for tests involving dispersion indicate that a lack of smooth earnings increases forecast dispersion. Table 4 also provides the results of estimating equation (6), which includes the interaction of credit ratings with earnings quality (CR*EQ). As in our prior test, the main effect of CR is positive and significant (0.0352, p<0.0001). Note that all the control variables' direction and significance are also consistent with the prior test. The CR*EQ interaction term is negative and significant (-1.4310, p<0.0001) indicating that the quality of reported earnings affects the influence that credit ratings have on forecast accuracy. Specifically, at higher levels of credit ratings, the negative association between earnings quality and financial analyst forecast accuracy is reduced. This result is consistent with our expectations regarding the moderating effect that earnings quality has on the relationship between credit ratings and financial analyst forecast accuracy and provides evidence for the moderating effect of earnings quality anticipated by Hypothesis 4. The results with DISP as the dependent variable similarly support our hypothesis regarding the moderating effect of earnings quality. For the test involving forecast dispersion, the CR*EQ interaction term is positive and significant (-0.1617, p<0.0001) indicating that the quality of reported earnings affects the influence that credit ratings have on forecast accuracy. Specifically, at higher level of credit ratings, the positive association between earnings quality and financial analyst forecast dispersion is reduced. This result is also consistent with our expectations for Hypothesis 4 regarding the moderating effect that earnings quality has on the relationship between credit ratings and financial analyst forecast accuracy. For each of our tests that do not include interaction variables in Table 4, the variance inflation factors for all of the independent variables are less than 4, indicating low levels of multicollinearity among the independent variables.6 We also test the null hypothesis that there is constant variance of the residuals using the White test for all models. We find that heteroscedasticity is present and we therefore calculate all significance levels for our tests using heteroscedasticity consistent standard errors. Additionally, we test for autocorrelation and find evidence that the residuals may not be independent. We therefore re-estimate our model correcting for autocorrelation and find similar results. Figure 1 Analyst Forecast Error by Credit Rating CCC-, CCC Credit Rating » Median AFEP — Mean AFEP Figure 1 provides further analysis of the relationship between credit ratings and analyst forecast accuracy. Figure 1 shows an increase in analyst forecast accuracy as the credit rating of a firm increases. The increase in accuracy is most pronounced from the lowest ratings which are "D, CCC-, CCC" which have a median AFEP value of -20.248 to "CCC+" ratings which have a median AFEP value of -3.4464.7 However, increases in 6 The largest variance inflation factors in these models occur among the industry control dummy variables with the largest value at 3.3. The variance inflation factor for all other independent variables is less than 2. 7 We combine firms with a "D", "CCC-", or a "CCC" rating due to the small number of firms that receive these ratings. Only a combined 77 of the 88,652 observations with a credit rating in our sample received one of these three ratings. In contrast, 452 firms received a "CCC+" rating. accuracy continue throughout each of the non-investment grade ratings to firms with a "BB+" rating which have a median AFEP value of -0.2039. There is much less variation in investment grade ratings as "BBB-" rated firms have a median AFEP value of -0.1860 and firms with a "AAA" rating have a median AFEP value of -0.0944 with no investment grade rated firm having a median AFEP value of more than -0.0661. While there is not as much improvement in forecast accuracy as credit ratings improve for investment grade firms, there is dramatic improvement in forecast accuracy among non-investment grade firms as creditratings increase.Overall, across allcredit ratings, Figure 1 shows that as credit ratings mcr ecse, thereisaclevi mfreasem fo recast aceuracy. 5. ADDITtONALANALYSES AND aEOSITaVSTYTtSOS In orderte teaUhe censitivityofrvreerulteto alternative measuaeforccetain variables, we perform four additional tests. First, we use smoothness as an alternative measurement of earningsquality.Second, weutilizecredit rating changes instead of the actual credit rating as an alternative specification. Third, we test for endogeneity using Granger causality tests. Last, wc replicate our teotinguoingalhernLaivespeciaaLtionsoeanalystfoarcastacruercy. 5.1. SmoorVnessaraMeasureoVEarningsQualiCy We replicate our results using smoothness as an alternative measure of earnings quality to test the consistency of reported earnings. As in prior literature (e.g. Leuz, Nanda, & Wysocki, 2003; Gat&Zlran0, a0c5),wecomLutermeai^lnessoeeaNnlngr rr foUovs: a (NI / ASSETS), t SMOOTH Jt =-— (6) J,t a (CFO / ASSETS) Jt where for firm j and at time t: a = standacL da^-eNlia^e^^rsthe most recent 5 year period NI = n art meame CFO = cash flow from operations ASSETS = total rasets. Because Mgtaf levalh of rnriance in earmn^ rclatipa to csrh flowi lndiraid that a firm's earningsarrmorcvolsrile,Pirhrrlevclstf SMONTH ihdicatelowee Sevels of. sarnings smoothncss.CoLreiSpIp, lonrrlevelsof vcelancein ldr nlnga reiativctocashflowsindi lite that earning orrmorecvnslitentThesefoaClirwerlevdc sf SMOOnts^nViPate Mghas levels ofearnn^c smoothnecs. Asconeistent roc.yinye vse v veslrableearnlnra attsibnte (Graham, Haevey, & R^'gope, fCCet. eiimr marl mampulcte earnmgs tca^ch^^^c c rmoh^^ earnings path. As is the case with overall earnings quality, this manipulation of earnings can increase the uncertainty of the forecastinaenvirOTmenl end increase the difficulty of the forecasting environment leading to a reduction in financial analyst forecast accuracy and more dispersion among financial analyst forecasts. Therefore, we expect that SMOOTH will have a negative effect on financial analyst accuracy or AFEP and a positive effect on DISP. _Table 5_ _Rating with AFEP and DISP including SMOOTH_ Panel A Panel B Dependent variable: AFEP Dependent variable: DISP Independent Variable Coefficient Coefficient Coefficient Coefficient Intercept 0.0272 0.5913 *** 0.0062 *** 0.0032 *** CR 0.0370 *** -0.0016 -0.0007 *** -0.0005 *** SMOOTH -0.0974 *** -0.4565 *** 0.0001 0.0020 *** CR*EQ 0.0258 *** -0.0001 *** FOLLOW1 9.1900 *** 9.2200 *** -0.0159 * -0.0160 * LOSS -1.2732 *** -1.2190 *** 0.0173 *** 0.0170 *** AEARN -19.2472 *** -18.9791 *** 0.2169 *** 0.2155 *** HORIZ1 -3.8200 *** -3.8000 *** 0.0013 0.0012 VOL 0.4150 0.2987 0.0703 *** 0.0709 *** IND Included Included Included Included YEAR Included Included Included Included Adjusted R2 0.3369 0.3434 0.4828 0.4845 n 88,652 88,652 88,652 88,652 *, **, *** indicate statistical significance utilizing heteroscedasticity consistent standard errors at the 10%, 5%, and 1% levels, respectively. 'These coefficients are multiplied by 1,000 for ease of interpretation. IND is an industry control using the I/B/E/S industry classification. YEAR is an annual control representing the year in which firm j's earnings are reported. All other variables are defined in Table 1. When substituting the SMOOTH variable for EQ in our equations (5) and (6), we find that, with the exception of the result for the VOL control, our results are consistent with Table 4. Our results in Table 5 demonstrate that firms with higher credit ratings and smoother earnings are associated with more accurate financial analyst forecasts. In Table 5, we also find some evidence that higher credit ratings and smoother earnings are associated with less forecast dispersion. While in our tests of equation (5) only higher credit ratings are associated with lower forecast dispersion, our tests of equation (6) show that both higher credit ratings and smoother earnings are both associated with less forecast dispersion. Additionally, we find that our tests regarding the interaction between credit ratings and earnings smoothness and its effect on accuracy and dispersion remain significant. Specifically, at higher level of credit ratings, both the negative association between earnings quality and financial analyst forecast accuracy and the positive association between earnings smoothness and financial analyst forecast dispersion are reduced. Taken together with our results from Table 4, our results when substituting SMOOTH for EQ in our equations (5) and (6), we find further evidence for Hypotheses 2, 3, and 4. 5.2. Change in Credit Rating In some cases, prior literature utilizes the change in credit rating to study external reports of a firm's financial position (Ederington & Goh, 1998); therefore we replicate our results utilizing this measure in place of the firm's current rating. _Table 6_ Change in rating with AFEP and DISP including EQ Panel A Panel B Dependent variable: AFEP Dependent variable: DISP Independent Variable Coefficient Coefficient Coefficient Coefficient Intercept 0.3454 *** 0.3501 *** -0.0020 *** -0.0019 *** ACR 0.2373 *** 0.2749 *** -0.0026 *** -0.0022 *** EQ 0.5205 0.5547 -0.0107 *** -0.0103 *** CR*EQ -1.8085 * -0.0214 ** FOLLOW1 13.6500 *** 13.5600 *** -0.0907 *** -0.0917 *** LOSS -1.2904 *** -1.2955 *** 0.0177 *** 0.0176 *** AEARN -19.1504 *** -19.1763 *** 0.2209 *** 0.2206 *** HORIZ1 -3.8500 *** -3.8500 *** 0.0018 ** 0.0018 ** VOL -3.1104 *** -3.0802 *** 0.0910 *** 0.0914 *** IND Included Included Included Included YEAR Included Included Included Included Adjusted R2 0.3381 0.3382 0.4833 0.4833 n 88,652 88,652 88,652 88,652 *, **, *** indicate statistical significance utilizing heteroscedasticity consistent standard errors at the 10%, 5%, and 1% levels, respectively. 'These coefficients are multiplied by 1,000 for ease of interpretation. IND is an industry control using the I/B/E/S industry classification. YEAR is an annual control representing the year in which firm j's earnings are reported. All other variables are defined in Table 1. Our results in Table 6 indicate that our results regarding the relationships between credit ratings, earnings quality and forecast accuracy as well as forecast dispersion are essentially unchanged. The significance level of the interaction variables decreases, but remains consistent. These results again confirm Hypotheses 2, 3, and 4 regarding the association between financial analysts' forecasts and firm credit ratings and earnings' attributes. 5.3. Endogeneity between Credit Ratings and Financial Analyst Forecast Performance As outlined above, we expect that firms with higher credit ratings allow for financial analysts to make their earnings forecast for a firm with more accuracy and less uncertainty. However, the possibility exists that financial analyst activity itself could have an effect on the credit ratings that are issued. For example, Cheng and Subramanyam (2008) find that a reduction in financial analysts following a firm leads to a decrease in credit ratings. Similarly, Mansi, Maxwell, and Miller (2011) find that an increase in financial analyst activity leads to a decrease in credit yields. Therefore, it is possible that an endogenous relationship exists between credit ratings and financial analyst forecast accuracy and dispersion. We examine this possibility by performing Granger causality tests to explore the connection between financial analyst activity and credit ratings. In untabulated results, we find results similar to those of Ederington and Goh (1998), in that Granger causality flows both ways. Our testing demonstrates significant results when regressing forecast accuracy on both prior period credit ratings and prior period forecast accuracy. At the same time, we also find significant results when regressing credit ratings on both prior period forecast accuracy and prior period credit ratings. While these results provide some evidence that credit ratings have an effect on performance of financial analysts, financial analyst activity also can have a simultaneous effect on the reports issued by credit rating agencies. 5.4. Alternative Measures of Analyst Forecast Accuracy We replicate our results using several alternate measures of forecast accuracy to test the sensitivity of our results to other measures of analyst forecast accuracy. First, we replicate our results using two alternate measures in the numerator. Prior literature has utilized a log transformation to reduce skewness in variables (Nikolaev 2010). Therefore, we begin our alternate measurement analysis by computing the natural log of the difference betweenthe analystforecastandthenetincomeof the firm (L/VAFjBP).8 Wereplicateour tests using ihio altsrnate meaauecment as follows: InilOOO*(FORECASTfTf-NI< t)) LNAFEP;; t = n-(-^-^ (-1) (7) l'J't PRICEst ( ) ( ) where for analyst i, firm j, and at time t: FORECASy =fore eosteb pet income Ni y fe yao rte tl toetlncome PyECy = ctoclk.rice. yonsisteno wit.nradshsw st oi. (2016)we oompute onalyst: forscost eseor utilizing tine aqcared deffercnoe between tiic anahyst forecai.. and the net: imcoms oCtl)2 fism CSQ-0—0). Theruforf, we replieate our eesuhs utlHomnthu following substitute measure: (FORECAST; j}—NI - t)2 ( ) SQAFEPU.t = 1-„rCE- „ (-100) (8) itcre foropelyst t,drmJ,anci attime t: FOCECASw = fore eosteb pet income M = oa (eor l^ed^^t income PRICi = atock price. e^^^l^^^git, dor intsrpcetuliuf curaeieu wf mu)lip)y ail elier foffceei f((ootqualktns (p taicl^ staCy dy l^t^O^, ductcs) oh SUs Oimlnottv) natives aeiu)iiyr by iCsk>) trensfotmationinroietiop7J ne(elhei eta amOijply lin aumetaten by OeOS). an) tCetttofe muliiplu bits forecast eiiroc by (-1). Additionally, we check theconsistency of our results by scaling the computation of the difference between the analyst forecast and the net income of the firm with earnings per share insteadofprice in thedenominator(Collins,HopAoo el,&McKeown, 1984).Utilizing earmngspershare to acalfanolfot rcowraoy,Ko compute forecast stror gEPSA^Plas followsi \FORECAST.t~;~}-Nlj t\ , EPSAFEPij; = ---- (-100) (9) where for analyst i, firm j, and at time t: FORnCAS A s=fore ratted nei income Nd = ae por trdwet income EPS = aeportede oanmgsp er share. Overall, when utilizing these three alternative specifications of analyst forecast accuracy, we find 1 argely s imilar r esults. For example, when testing our results utilizing the natural log of the differencesbetween analyst forecasts and earnings (LNAFEP) as a part of equation(5), inuntabuiated re eultc tshecoefficicrrtofCRce m aine positive ands^nifiaertat onep ercend inds aatingtioe0 higheeacc ur acy in analysts' forecasts of earnings is associated with firms that have higher credit ratings. Further, the coefficient of EQ remains positive and significant, but the level of significance drops from five percent to ten percent. When utilizing the alternative specification SQAFEP that determines analyst forecast error by calculating the squared difference between the forecast and the actual earnings, in untabulated results we find that the correlation between SQAFEP and CR is positive and significant. However, when SQAFEP is included as the dependent variable in equation (5), the coefficient on CR is negative and significant. In further testing, after winsorizing each of the continuous variables at 5%, and testing again SQAFEP as the dependent variable in equation (5), the coefficient on CR is again positive and significant at 1% indicating that outliers are the cause of our spurious results in the initial test of SQAFEP in equation (5). Finally, when replicating our results with earnings per share as an alternate to share price in the denominator of the measure as described above (EPSAFEP) in equation (5), in untabulated results the results are consistent with our main tests as we find that the coefficient on CR is positive and significant at 1% and the coefficient on EQ is positive and significant at 5%. 6. SUMMARY AND CONCLUSION Using models of forecast performance from prior literature, we examine the relationship of credit ratings with the accuracy of financial analysts' forecasts and the dispersion of forecasts. Controlling for factors that are known in the literature to introduce task complexity into the forecasting environment, we hypothesize that high credit ratings are associated with more accurate and less dispersed earnings forecasts. This study provides several new results. First, we find that the existence of a firm's credit ratings during the forecasting horizon has a significant impact on the accuracy and dispersion patterns of financial analysts' earnings forecasts. Second, we find that higher credit ratings are associated with more accurate forecasts and less dispersed forecast patterns. Next, we find that the level of a firm's earnings quality moderates the effect that credit ratings have on analyst performance. This particular result suggests that the coexistence of high credit ratings and high quality earnings reports provides a less complex and more forecast-friendly environment. In additional testing we use three alternate specifications of forecast accuracy to repeat our tests. The results of these alternate tests are consistent with our primary findings. This study is motivated by making contributions that provide 1) analysts with confirmation of the usefulness of credit ratings in assessing future earnings, 2) regulators with further evidence of the importance of high quality earnings reports, 3) investors with the knowledge that future earnings forecasts are likely more accurate for a firm with higher credit ratings, and 4) researchers with additional insights of the determinants of financial analyst forecasting performance. Additionally, these results also add to the body of literature that illustrates how task complexity hinders the predictive value of earnings reports. Our results allow for related future research in several ways. First, future studies could explore the link between credit ratings and changes in financial analysts' performance based on whether it is classified as investment grade, speculative grade, or other various subcategories. Second, future studies could utilize measures of relative analyst forecast accuracy to determine whether there is an interactive effect between credit ratings and specific analyst characteristics, such as analyst experience, on their performance. Finally, extensions of this research could examine whether and to what degree the level of credit ratings interact with other variables known in the literature to have a significant impact on forecast accuracy and dispersion. Our research is subject to certain limitations. As we have reported in our testing, while we find that credit ratings have an effect on financial analyst performance, there is also a simultaneous effect that financial analysts' earnings forecasts may have on credit ratings themselves. Additionally, the Compustat database utilized in this study includes Standard & Poor's credit ratings exclusively and thus future research could duplicate our testing with data from all rating agencies. Finally, although we use controls in our testing that are common for this area in the literature, as is the case in all empirical research our models may be misspecified due to the possibility of omitted control variables. REFERENCE LIST Ashbaugh-Skaife, H., Collins, D.W. & LaFond, R. (2006). The effect of corporate governance on firms' credit ratings. Journal of Accounting and Economics, 42 (1/2), 203-43. Bae, K.H., Tan, H. & Welker, M. (2008). International GAAP differences: The impact on foreign analysts. The Accounting Review, 83(3), 593-628. Bae, K.H., Purda, L., Welker, M. & Zhong, L. (2013). Credit rating initiation and accounting quality for emerging market firms. Journal of International Business Studies, 44(3), 216-34. Barniv, R., Hope, O., Myring, M.J. & Thomas, W.B. (2009). Do analysts practice what they preach and should investors listen? Effects of recent regulations. The Accounting Review, 84(4), 1015-39. Barron, O.E. & Stuerke, P.S. (1998). Dispersion in analysts' earnings forecasts as a measure of uncertainty. Journal of Accounting, Auditing and Finance, 13(3), 245-70. Barth, M.E., Elliott, J.A. & Finn, M.W. (1999). Market rewards associated with patterns of increasing earnings. Journal of Accounting Research, 37(2), 387-413. Barton, J., Hansen, T.B. & Pownall, G. (2010). Which performance measures do investors around the world value the most--and why? The Accounting Review, 85(3), 753-89. Basu, S., LeeSeok, H. & Ching-Lih, J. (1998). International variation in accounting measurement rules and analysts' earnings forecast errors. Journal of Business Finance and Accounting, 25(9/10), 1207-47. Bilinski, P. (2014). Do analysts disclose cash flow forecasts with earnings estimates when earnings quality is low? Journal of Business Finance and Accounting, 41(3/4), 401-34. Bradshaw, M. T., Plumlee, M. A., Whipple, B. C. & Yohn, T. L. (2016). The impact of earnings announcements on a firm's information environment. Working paper. Brown, L. (2001). A temporal analysis of earnings surprises: Profits versus losses. Journal of Accounting Research, 39(2), 221-42. Burgstahler, D. & Eames, M. (2006). Management of earnings and analysts' forecasts to achieve zero and small positive earnings surprises. Journal of Business Finance and Accounting, 33(5/6), 633-52. Byard, D., & Shaw, K.W. (2003). Corporate disclosure quality and properties of analysts' information environment. Journal of Accounting, Auditing and Finance, 18(3), 355-78. Cheng, M., & Subramanyam, K.R. (2008). Analyst following and credit ratings. Contemporary Accounting Research, 25(4), 1007-43. Clement, M.B. (1999). Analyst forecast accuracy: Do ability, resources, and portfolio complexity matter? Journal of Accounting and Economics, 27(3), 285-303. Clement, M.B. & Tse, S.V. (2003). Do investors respond to analysts' forecast revisions as if forecast accuracy is all that matters? The Accounting Review, 78(1), 227-49. Cohen, D.A., Dey, A. & Lys, T.Z. (2008). Real and accrual-based earnings management in the pre- and post-Sarbanes-Oxley periods. The Accounting Review, 83(3), 757-787. Collins, W. A., Hopwood, W. S. & McKeown, J. C. (1984). The predictability of interim earnings over alternative quarters. Journal of Accounting Research, 22(2), 467-79. Dechow, P. & Dichev, I. (2002). The quality of accruals and earnings: The role of accrual estimation errors. The Accounting Review, 77(4), 35-59. Dechow, P., Ge, W. & Schrand, C. (2010). Understanding earnings quality: A review of the proxies, their determinants, and their consequences. Journal of Accounting and Economics, 50(2/3), 344-401. Dechow, P., & Skinner, D.J. (2000). Earnings management: Reconciling the views of accounting academics, practitioners, and regulators. Accounting Horizons, 14(2), 235-50. Demerjian, P.R., Lewis-Western, M.F. & McVay, S.E. (2015). Earnings smoothing: For good or evil? Working paper. Dhaliwal, D.S., Radhakrishnan, S., Tsang, A. & Yong George, Y. (2012). Nonfinancial disclosure and analyst forecast accuracy: International evidence on corporate social responsibility disclosure. The Accounting Review, 87(3), 723-59. Diether, K.B., Malloy, C.J. & Scherbina, A. (2002). Differences of opinion and the cross section of stock returns. The Journal of Finance, 57(5), 2113-41. Drake, M.S. & Myers, L.A. (2011). Analysts accrual-related over-optimism: Do analyst characteristics play a role? Review of Accounting Studies, 16(1), 59-88. Duru, A. & Reeb, D. (2002). International diversification and analysts' forecast accuracy and bias. The Accounting Review, 77(2), 415-33. Ederington, L. & Goh, J. (1998). Bond rating agencies and stock analysts: Who knows what when? Journal of Financial and Quantitative Analysis, 33(4), 569-85. Elliott, J. A. & Hanna, J.D. (1996). Repeated accounting write-offs and the information content of earnings. Journal of Accounting Research, 34(3), 135-55. Francis, J., LaFond, R., Olsson, P. & Schipper, K. (2004). Costs of equity and earnings attributes. The Accounting Review, 79(4), 967-1010. Francis, J., LaFond, R., Olsson, P. & Schipper, K. (2005). The market pricing of accruals quality. Journal of Accounting and Economics, 39(2), 295-327. Frost, C.A. (2007). Credit rating agencies in capital markets: A review of research evidence on selected criticisms of the agencies. Journal of Accounting, Auditing and Finance, 22(3), 469-92. Gao, L. & Zhang, J. H. (2015). Firms' earnings smoothing, corporate social responsibility, and valuation. Journal of Corporate Finance, 32(1), 108-27. Glaum, M., Baetge, J., Grothe, A. & Oberdorster, T. (2013). Introduction of international accounting standards, disclosure quality and accuracy of analysts' earnings forecasts. European Accounting Review, 22(1), 79-116. Graham, J.R., Harvey, C.R. & Rajgopal, S. (2005). The economic implications of corporate financial reporting. Journal of Accounting and Economics, 40(1), 3-73. Gray, S., Mirkovic, A. & Ragunathan, V. (2006). The determinants of credit ratings: Australian evidence. Australian Journal of Management, 31(2), 333-54. Guan, Y., Lu, H. & Wong, M.F. (2012). Conflict-of-interest reforms and investment bank analysts' research biases. Journal of Accounting, Auditing and Finance, 27(4), 443-70. Hand, J.M., Holthausen, R.W. & Leftwich, R.W. (1992). The effect of bond rating agency announcements on bond and stock prices. Journal of Finance, 47(2), 733-52. Hong, H., Kubik, J. & Solomon, A. (2000). Security analysts' career concerns and herding of earnings forecasts. RAND Journal of Economics, 31(1), 121-44. Hwang, L.S., Jan, C.L. & Basu, S. (1996). Loss firms and analysts' earnings forecast errors. The Journal of Financial Statement Analysis, 1(2), 18-30. Imhoff, E.A., & Lobo, G.J. (1992). The effect of ex ante earnings uncertainty on earnings response coefficients. The Accounting Review, 67(2), 427-39. Jacob, J., Lys, T. & Neale, M. (1999). Expertise in forecasting performance of security analysts. Journal of Accounting and Economics, 28(1), 51-82. Jacob, J., Rock, S. & Weber, D.P. (2008). Do non-investment bank analysts make better earnings forecasts? Journal of Accounting, Auditing and Finance, 23(1), 23-61. Jorion, P., Zhu, L. & Charles, S. (2005). Information effects of Regulation FD: Evidence from rating agencies. Journal of Financial Economics, 76(2), 309-30. Kinney, W., Burgstahler, D. & Martin, R. (2002). Earnings surprise "materiality" as measured by stock returns. Journal of Accounting Research, 40(5), 1297-329. Kliger, D. & Sarig, O. (2000). The information value of bond ratings. Journal of Finance, 55(6), 2879-902. Kross, W., Ro, B. & Schroeder, D. (1990). Earnings expectations: The analysts' information advantage. The Accounting Review, 65(2), 461-76. Lang, M.H. & Lundholm, R.J. (1996). Corporate disclosure policy and analyst behavior. The Accounting Review, 71(4), 467-92. Lehavy, R., Li, F. & Merkley, K. (2011). The effect of annual report readability on analyst following and the properties of their earnings forecasts. The Accounting Review, 86(3), 1087-115. Leuz, C., Nanda, D. & Wysocki, P.D. (2003). Earnings management and investor protection: An international comparison. Journal of Financial Economics, 69(3), 505-27. Liang, L. & Riedl, E. J. (2014). The effect of fair value versus historical cost reporting model on analyst forecast accuracy. The Accounting Review, 89(3), 1151-77. Lim, T. (2001). Rationality and analysts' forecast bias. Journal of Finance, 56(1), 369-85. Lobo, G., Song, M. & Stanford, M. (2012). Accruals quality and analyst coverage. Journal of Banking and Finance, 36(2), 497-508. Mansi, S. A., Maxwell, W. F. & Miller, D. P. (2011). Analyst forecast characteristics and the cost of debt. Review of Accounting Studies, 16(1), 116-42. Matsumoto, D. (2002). Management's incentives to avoid negative earnings surprises. The Accounting Review, 77(3), 483-514. Mauri, A. J., Lin, J. & Neiva DeFigueiredo, J. (2013). The influence of strategic patterns of internationalization on the accuracy and bias of earnings forecasts by financial analysts. International Business Review, 22(4), 725-35. McEwen, R.A. & Hunton, J.E. (1999). Is analyst forecast accuracy associated with accounting information use? Accounting Horizons, 13(1), 1-16. Mensah, Y.M. & Yang, R. (2008). An empirical evaluation of analysts' herding behavior following Regulation Fair Disclosure. Journal of Accounting and Public Policy, 27(4), 317-38. Mikhail, M.B., Walther, B.R. & Willis, R.H. (1997). Do security analysts improve their performance with experience? Journal of Accounting Research, 35(Supp.), 131-57. Mizik, N. (2010). The theory and practice of myopic management. Journal of Marketing Research, 47(4), 594-611. Moehrle, S.R. (2002). Do firms use restructuring charge reversals to meet earnings targets? The Accounting Review, 77(2), 397-413. Murphy, K.J. (2001). Performance standards in incentive contracts. Journal of Accounting and Economics, 30(3), 245-78. Nikolaev, V. V. (2010). Debt covenants and accounting conservatism. Journal of Accounting Research, 48(1), 137-76. Norden, L., & Weber, M. (2004). Informational efficiency of credit default swap and stock markets: The impact of credit rating announcements. Journal of Banking and Finance, 28(11), 2813-43. Payne, J. & Robb, S. (2000). Earnings management: The effect of ex ante earnings uncertainty. Journal of Accounting, Auditing and Finance, 15(4), 371-92. Partnoy, F. (1999). The Siskel and Ebert of financial markets? Two thumbs down for the credit rating agencies. Washington University Law Quarterly, 77(3), 619-714. Pottier, S.W. & Sommer, D.W. (1999). Property-liability insurer financial strength ratings: Differences across rating agencies. Journal of Risk and Insurance, 66(4), 621-42. Skinner, D.J. & Sloan, R.G. (2002). Earnings surprises, growth expectations, and stock returns or don't let an earnings torpedo sink your portfolio. Review of Accounting Studies, 7(2), 289-312. Tan, H., Wang, S. & Welker, M. (2011). Analyst following and forecast accuracy after mandated IFRS adoptions. Journal of Accounting Research, 49(5), 1307-57. WHAT REALLY DEFINES THE PERFORMANCE IN HOTEL INDUSTRY? MANAGERS' PERSPECTIVE USING DELPHI METHOD Valentina Božič1 Ljubica Kneževic Cvelbar2 Received: April 6, 2017 Accepted: June 5, 2018 ABSTRACT: This paper brings together empirical studies in hospitality literature focusing on understanding the drivers of hotel performance and combines them with experts' opinions on the most important drivers of performance in the hospitality industry. The first step focused on studying more than 60 papers available in hospitality literature focusing on hotel performance. The second step involved the Dephi method on a pool of 10 experts from the hospitality industry in order to explore their opinion on major drivers of hotel performance. The results showed that literature is mainly focusing on examining the impact of HRM practices, organizational culture, ICT, brand equity, environmental practices and hotel facilities on performance. However, experts did not find those drivers to be the crucial factors of hotel performance. They emphasized the importance of location, product segmentation, company flexibility and cooperation between tourist service providers. Those drivers have not caught the attention of academia so far and they represent a potential new avenue for future hospitality research on understanding hotel performance. Both literature and experts found that market orientation, customer satisfaction, service quality and business processes are important drivers of hotel performance that require further research and examination. Keywords: Delphi method, performance, hotels, drivers of performance JEL: L83 DOI: 10.15458/85451.70 1. INTRODUCTION The hospitality industry has for decades been striving to understand the major drivers of hotel performance. So far literature offered many fragmented studies examining specific resources or capabilities that drive hotel performance. Theoretical background behind this body of literature is the Resource Based Theory (RBT) proposed by Barney in 1990. This theory postulates the basis for defining sustainable competitive advantages of any firm. Resources become a competitive advantage if they are valuable, rare, inimitable and non-substitutable (Barney, 1991). Further development of RBT made a distinction 1 Corresponding author, University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, PhD Candidate, e-mail: valentina@navis.agency 2 University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: ljubica.knezevic@ef.uni-lj.si between firms' resources. Collis (1994) classified recourses as tangible and intangible. Tangible resources are physical and financial assets of the firm and represent the value of the financial capital. Intangible resources are non-physical and they are rarely part of firms financial statements. Amit and Schoemaker (1993) recognized that capabilities are also a source of a firm's competitive advantage and defined them as routines based on skills and knowledge that can drive the firm's success. Overall, general management literature on RBT focused on defining and clarifying tangible and intangible resources and capabilities that drive firms' performance. Theoretical papers in management literature worked on clarifying the general theoretical concepts that highlighted the nature of firms' resources and capabilities. As for empirical papers, they were operationalizing and testing those concepts in practice. The major challenges in empirical studies were: 1) measurement issues (how can abstract concepts be measured?); 2) the scope of concepts that can be included in one study (the concepts are broad and a single study could not focus on all of them); and 3) weak boundaries between different concepts (how can we distinguish between intangible assets and capabilities and where is the line between them in practice?). Those challenges caused a gap between theoretical and empirical papers since the concepts defined in the theory were hard to test and measure empirically. In hospitality industry there are no theoretical papers dealing with drivers of hotel performance. Most of the literature is based on empirical studies using general management RBT literature and empirical research in line with generally accepted management concepts of assets and capabilities. In general, intangible assets and their relation to performance attracted the most attention in experts in the field. Those papers tested the impact of HRM practices, brand equity, information communication technologies (ICT), social capital, environmental policies, employees and managers' know-how, and their impact on hotel performance (Božič and Kneževic Cvelbar, 2016). Most of the studies in hospitality literature were focusing on one single or a few drivers of hotel performance, which were chosen based on the subject of the researcher's interest or general theoretical approval of the concept's relevance. The majority of those studies showed that there is a positive correlation between the driver(s) and hotel performance. Yet they failed to determine which drivers are the most important and relevant. This is an ultimate question for practitioners and hotel manages. Knowing what is positively impacting hotel performance is good, but knowing the major drivers of hotel performance is crucial. This paper is focusing on identifying the major drivers of hotel performance. It first presents an in-depth overview of the hospitality literature in order to indicate which drivers have been researched so far. Based on literature review, we identified the list of 30 drivers of performance that were used in academic research so far. A total of three rounds of testing were performed using the Delphi methodology. In each round experts were asked to evaluate certain drivers of hotel performance that had been recognized in hospitality literature in the basis of their perception of their impact on hotel performance. The list was consequently reduced in line with the previous round of results. The participants were also invited to suggest additional drivers of hotel performance. The research concluded after the third round since it yielded no new information. The final results provided a list of nine major drivers of hotel performance. Furthermore, the major drivers of performance were investigated in the literature and compared with the results of the Delphi study. The results are presented and discussed in this paper. 2. OVERVIEW OF PREVIOUS STUDIES AND RECOGNITION OF EXISTING DRIVERS OF PERFORMANCE Studies in the hospitality industry were following general trends in management literature, which means they revolved around testing the impact of tangible and intangible assets and capabilities on hotel performance. An extensive search for relevant literature for the purpose of this paper found 60 papers published in the last twenty years - testing the drivers of performance in the hospitality industry. The majority of those papers were published in the International Journal of Hospitality Management that was and is still supporting research on hotel performance. Those papers helped identify 30 major drives of hotel performance that managed to attract the attention of academics. The papers are listed and presented in Table 1. In line with the selected literature, firms' sustainable competitive advantages were classified as tangible assets, intangible assets, and capabilities. As defined above tangible assets denote a firm's capital translated into its assets (land, buildings, equipment etc.) and financial funds (cash and other financial assets). Interestingly, hospitality studies were not particularly interested in studying the impact of tangible assets on firms' performance. The reason for this could be the fact that tangible assets are necessary and obvious drivers of hotel success. Studies in this area were unified in finding that tangible assets, including hotel facilities, location and financial assets, are positively related to hotel performance. The line between intangible assets and capabilities is rather thin. There is no general consensus in literature as to which competitive advantages are intangible assets and which are, in fact, capabilities. Therefore classifying a specific sustainable competitive advantage as an intangible asset or a capability is arbitrary. This paper follows the simple logic proposed by Hall (1992), which says: "intangible assets are something that a firm has, while capabilities are something that a firm does" (pg. 136). In line with Hall (1992), the definition and the existing empirical studies in hospitality industry means that intangible assets can be grouped into four general categories as a firm's organizational, human, marketing and environmental assets. ■ Organizational assets are understood as the firm's culture, organizational structure, management philosophy, available informational technology, service quality, social capital, and the corporate social responsibility policy. Overall organisational assets are the most researched drivers of success in the hospitality industry. Studies are generally conclusive - organisational assets can be translated into firms' sustain competitive advantages leading to their success. ■ Another frequently researched area relates to human assets including management and employee competences, HRM practices, employees' attitudes, and employee satisfaction, innovativeness and loyalty. As expected, human assets positively correlate with hotel performance. Research in hospitality is very focused on employees, but it neglects management-related sustainable competitive advantages as performance drivers. This is a research field worth of exploring in the future. ■ Marketing capital includes brand equity, customer satisfaction, customer loyalty and direct distribution channels. As in the case of organisational and human capital, marketing capital is also positively related to firms' performance. Relevant literature has put the most effort on exploring customer satisfaction and its relation to performance. Distribution channels as part of the marketing capital are relatively underexplored, which is surprising due to the digital transformation of the industry that has been happening in the last decade. In addition, the value of contractual relations with business customers as a part of the sales policy has so far not been explored in hospitality studies. ■ Environmental capital is still not considered as a very important research area in the hospitality industry. Although sustainable development is one of the research mainstreams in tourism, the relation between environmental capital and hotel performance failed to attract significant academic interest. Very few studies connected basic and advanced environmental policies with hotel performance, but they generally found a positive relation between the two. More research in this area is expected to emerge in the future. Capabilities are prominent sources of firms' success and, according to RBT, they should represent the leading source of sustainable competitive advantages. The major obstacle in the research is operationalisation and measurement of firms' capabilities. Hospitality research in general neglected such capabilities as important sustainable competitive advantages of firms. Research in this area is still scarce. In general, capabilities can be divided into operational and dynamic. Operational capabilities are firms' routines and processes that enable them to perform activities in the long-term and ongoing basis, while dynamic capabilities relate to firms' ability to adapt to changes coming from the environment. Research in operational capabilities in the hospitality industry includes firms' relations with partners and business processes, while research in dynamic capabilities includes marketing orientation, knowledge sharing and entrepreneurship orientation. Summary of the literature review on drivers of hotel performance is presented in table 1. Table 1: Drivers of performance in hospitality literature Short description Authors and year Relation with performance performance Drivers based on tangible assets Hotel facilities Buildings, equipment Chu & Choi, 2000 (+) Lenidou, Leonidou, Fotiadis & Zeriti, 2013 (+) Kim, Cho & Brymer, 2013 (+) Lado-Sestayo, Otero-González, Vivel- Búa & Martorell-Cunill, 2016 (+) Location Physical location - land Lado-Sestayo et. al., 2016 (+) Financial assets Cash and other financial funds Lenidou et. al., 2013 (+) Drivers based on intangible assets Brand equity Brand loyalty, awareness, Prasad & Dev, 2000 (+) perceived quality, brand image Kim & Kim, 2005 (+) Rudez & Mihalic, 2007 (+) O'Neill & Carlbäck, 2011 (/) Employee loyalty Employee's sense of belonging Kim & Brymer, 2011 (+) and commitment Al-Rafaie, 2015 (+) Employee Working conditions, teamwork Chi & Gursoy, 2009 (/) satisfaction and cooperation, relationship Naseem, Sheikh & Malik, 2011 (+) with supervisors, recognition Al-Rafaie, 2015 (+) and awards Employee Qualifications, experience, Rudez & Mihalic, 2007 (+) competencies knowledge development, Kim, Kim, Park, Lee & Jee, 2012 (+) knowledge sharing between Nieves, Quintana & Osorio, 2014 (+) employees Employees' Overall satisfaction, pride, Sharpley & Forster, 2003 (+) attitudes towards consistency and devotion Rudez & Mihalic, 2007 (+) work Employee Creativity, innovative ideas of Rudez & Mihalic, 2007 (+) innovativeness employees Nieves, Quintana & Osorio, 2014 (+) HRM practices Recruitment and selection, Hoque, 1999 (+) manpower planning, Cho, Woods, Jang & Erdem, 2006 (+) job design, training and Chand & Katou, 2007 (+) development, pay system Chand, 2010 (+) Ahmad, Solnet & Scott, 2010 (/) Al-Rafaie, 2015 (+) Ruzic, 2015 (+) Managerial Analysis, strategic Kay & Russette, 2000 (+) competencies management, problem solving, Chung-Herrera, Enz & Lankau, 2003 (+) leadership, creativity, crisis Jeou-Shyan, Hsuan, Chih-Hsing, Lin & management, attitude, self Chang-Yen, 2011 (/) management Wu & Chen, 2015 (+) Management Empowered employees, Rudez & Mihalic, 2007 (+) philosophy customers come first, stimulated staff Team culture Support of knowledge, Hu, Horng & Sun, 2009 (+) information sharing, Hussain, Kronar & Ali, 2016 (+) coordination, meetings, pre- designed work plans and processes Organisational Atmosphere, support of Kemp & Dwayer, 2001 (+) culture knowledge and communication Sorensen, 2002 (/) Rudez & Mihalic, 2007 (+) Asree, Zain & Rizal Razalli, 2010 (+) Zoghbi-Manrique-de-Lara & Ting- Ding, 2016 (/) Customer loyalty Attitude and customers loyalty Kandampully & Suhartanto, 2000 (+) behaviour Al-Rafaie, 2015 (+) Kim, Voght & Knutson, 2015 (/) Customer Degree of satisfaction with the Wilkins, Merrilees & Herington, 2007 (+) satisfaction services, price, location and Rudez & Mihalic, 2007 (+) amenities Chi & Gursoy, 2009 (+) Assaf & Magnini, 2012 (+) Sun & Kim, 2013 (+) Kim, Cho & Brymer, 2013 (+) Al-Rafaie, 2015 (+) Kim, Voght & Knutson, 2015 (/) Assaf, Josiassen, Cvelbar & Woo, 2015 (+) Service quality Tangibility, reliability, Herrington & Akehurst, 1996 (+) responsiveness, assurance, Chu & Choi, 2000 (+) empathy Claver, Jose, Tari & Pereira, 2006 (+) Al-Rafaie, 2015 (+) Wang, Chen & Chen, 2012 (+) Pereira-Moliner, Claver-Cortes, Molina-Azorin & Tari, 2012 (+) Molina-Azorin, Tari, Pereira-Moliner, Jopez-Gamero & Pertusa-Ortega, 2015 (+) Social capital Passion to achieve common Kim, Kim, Park, Lee & Jee, 2012 (+) goals and vision, close Terry, Kim, Lee, Paek & Lee, 2013 (+) relationships between work Dai, Mao, Zhao & Mattila, 2015 (+) colleagues, cooperation between departments Direct Online marketing, direct mail, Rudez & Mihalic, 2007 (+) distribution mobile marketing, call-centres channels Information IT for front-office and Rudez & Mihalic, 2007 (+) technology (IT) bookings, databases, Sirirak, Islam & Ba Khang, 2011 (+) management information Mihalic & Buhalis, 2013 (/) system, customer relationship Oltean, Gabor and ConJiu, 2014 (/) management applications Mihalic, Garbin Pranicevic & Arneric, 2015 (+) Melian-Gonzales & Bulchand- Gidumal, 2016 (+) Organisational Type of structure: mechanistic, Jogaratnam & Ching-Yick Tse, 2006 (/) structure organic 0gaard, Marnburg & Larsen, 2008 (/) Tavitiyaman, Qiu Zhang & Qu, 2012 (/) Corporate Ownership structure, board, Knezevic Cvelbar & Mihalic, 2007 (+) governance CEO or general manager's Assaf & Cvelbar, 2011 (+) characteristics Xiao, O'Neill & Mattila, 2012 (+) Jarboui, Guetat & Boujelbéne, 2015 (+) Basic Ecological product usage, Molina-Azorin, Claver-Cortés, environmental reduction in the use of Pereira-Moliner & Tari, 2009 (+) practices dangerous products, energy Tari, Claver-Cortés, Pereira- and water saving practices, Moliner & Molina-Azorin, 2010 (+) selective waste collection Pereira-Moliner, Claver-Cortés, Molina-Azorin & Tari, 2012 (+) Advanced Employee environmental Molina-Azorin, Claver-Cortés, environmental trainings and initiatives, Pereira-Moliner & Tari, 2009 (+) practices ecological marketing Tari, Claver-Cortés, Pereira- campaigns and events, long- Moliner & Molina-Azorin, 2010 (+) term environmental policies Pereira-Moliner, Claver-Cortés, and goals Molina-Azorin & Tari, 2012 (+) Leonidou, Leonidou, Fotiadis & Zeriti, 2013 (+) Corporate social CRS values, hiring locals, Kang, Lee & Huh, 2010 (/) responsibility ethnical and overseas De Grosbois, 2012 (/) (CSR) practices employees, environmental Garay & Font, 2012 (+) savings (recycling, reducing Assaf & Josiassen & Cvelbar, 2012 (+) energy costs, reusing towels, Benavides-Velasco, Quintana-Garcia & linen etc.) Marchante-Lara, 2014 (+) Fu, Ye & Law, 2014 (+) Drivers based on capabilities Relationships with Relations with customers, Rudez & Mihalic, 2007 (+) commercial and suppliers and other partners Kim, Kim, Park, Lee & Jee, 2012 (+) other partners Business Hotel standards and Rudez & Mihalic, 2007 (+) processes procedures, service Claver-Cortes, Pereira-Moliner, Tari & performance, customer Molina-Azorin, 2008 (+) complaint solving procedures, Wang, Chen & Chen, 2012 (+) innovative ideas, continuous Benavides-Velasco, Quintana-Garcia & process improvement Merchante-Lara, 2014 (+) Knowledge Knowledge sharing between Sristava, Bartol & Locke, 2006 (+) sharing teams, willingness to learn and Hu, Horng & Sun, 2009 (+) help others Terry, Kim, Lee, Paek & Lee, 2013 (+) Hussain, Konar & Ali, 2016 (+) Market Orientation to customers, Gray, Matear & Matheson, 2000 (+) orientation competitors, seeking profitable Matear, Osborne, Garrett & Gray, 2002 (+) customers and products, Barros & Dieke, 2008 (+) responsiveness to market Assaf & Cvelbar, 2011 (+) changes Wang, Chen & Chen, 2012 (/) Josiassen & Assaf & Knezevic Cvelbar, 2014 (+) Vega-Vázquez, Cossio-Silva & Revilla- Camacho, 2016 (+) Entrepreneurial Innovativeness, risk-taking, Jogaratnam & Ching-Yick Tse, 2006 (+) orientation pro-activeness, autonomy Vega-Vázquez, Cossio-Silva and (+) Revilla-Camacho, 2016 Hernández-Perlines, 2016 (+) (+) - positive impact on performance; (/) - positive impact on performance is not confirmed. 3. METHODOLOGY So far, literature identified many drivers of hotel success. However, very few of them attempted to determine the relevant competitive advantages in the hospitality industry. This paper is therefore addressing this important question. Qualitative research, i.e. the Delphi survey, was used to that end. The Delphi survey is a group research technique that collects opinions of anonymous experts from a certain area and transforms them through a series of rounds into a common group consensus (McKenna, 1994). Anonymous group experts receive a questionnaire in each round, complete it and send it back to the group facilitator. The facilitator collects all questionnaires, summarizes the answers and sends them back to the members of the group. The members again reconsider their answers based on the summarized group answers from the previous round and complete the questionnaire once again. The rounds finish when there is no further progress in the opinions of experts that would change the group's common result. The main limitation of the Delphi method is the subjective evaluation of the respondents and impact of the panel on respondent opinion (respondent could evaluate a specific item higher or lower in the second or third round based on the results from the previous round of research). Delphi is also narrowly applicable to few specific setting, usually in the case when policy recommendation has to be set. This was our case, where we needed guidance on defining relevant drivers of performance in hospitality industry. This method is also very lengthy and complex to conduct. In our case, the panel included 10 hospitality experts from Slovenia and Croatia. The group of experts was carefully selected to include three hotel managers, three hotel general managers and four representatives of academia, all with profound knowledge and great interest in hotel performance. The panel experts' general characteristics are presented in Table 2 below. Table 2: Respondents' characteristics Respondent Position Experience in the industry Age Gender Country of origin No. 1 Hospitality & Tourism Professor 25 55 female Slovenia No. 2 CEO 11 36 male Slovenia No. 3 Hospitality & Tourism Professor 22 46 female Croatia No. 4 CEO 12 39 male Slovenia No. 5 Hospitality & Tourism Professor 19 45 female Slovenia No. 6 Hotel Manager 27 53 male Slovenia No. 7 Hotel Manager 15 35 female Slovenia No. 8 CEO 20 42 male Croatia No. 9 Hospitality & Tourism Professor 7 35 female Croatia No. 10 Hotel Manager 12 39 male Croatia The questionnaire was devised in-line with relevant literature. It included 30 recognised and significant drivers of hotel performance thus far investigated in hospitality research. The panel members' role was to evaluate the importance of each driver of hotel performance on the scale of1 to 7 (1 = not at all important impact, 7 = extremely important impact on hotel performance). They were also invited to contribute by providing their recommendations and proposing additional drivers of performance that were in their opinion neglected in hospitality research. Three rounds of evaluations were performed. In the third and final round the experts confirmed the results from the second round and did not offer new insights. The research took place from September to December 2016. 3. DELPHI ANALYSIS RESULTS The analysis of the Delphi results requires a basic statistical analysis including mean averages, frequencies and ranking. The evaluations of the experts from round 1 are summarized in Table 3. The average scores are distributed between 6.3 (the highest average score) and 4.3 (the lowest average score). The drivers that were found to have the strongest impact on hotel performance were: location, market orientation, service quality (average score of 6.3) as well as customer satisfaction and business processes (average score of 6.1). The drivers with the lowest level of importance on hotel performance appeared to be advanced and basic environmental practices, organisational structure and organisational culture (all received an average score of 4.8 or less). The experts suggested that product development, cooperation, investment management and flexibility should also be considered as important drivers of hotel performance. All four additional drivers were included in round 2 of panel evaluation. The results of round 1 of the Delphi study are presented in Table 3. Table 3: Delphi round 1 results Mean values % of respondents that rated the % of respondents that rated the % of respondents that rated the Indicator importance as high (scores 6 or 7) importance as moderate or neutral (scores 5, 4 or 3) importance as low or insignificant (scores 2, 1 or 0) Location 6.3 80% 20% 0% Market orientation 6.3 80% 20% 0% Service quality 6.3 80% 20% 0% Customer satisfaction 6.1 80% 20% 0% Business processes 6.1 70% 30% 0% Management philosophy 6.0 70% 30% 0% Managerial competencies S.9 70% 30% 0% Employee competencies S.9 70% 30% 0% Hotel facilities S.7 S0% 50% 0% Information Technology (IT) S.7 S0% 50% 0% Human resource management practices (HRM) S.6 60% 40% 0% Knowledge sharing S.6 S0% 50% 0% Social capital 5.6 50% 50% 0% Employee attitudes towards work 5.6 50% 50% 0% Relationship with commercial parties and other partners 5.5 50% 50% 0% Team culture 5.5 50% 50% 0% Corporate governance 5.3 70% 30% 0% Employee satisfaction 5.3 40% 60% 0% Employee innovativeness 5.2 50% 50% 0% Entrepreneurial orientation 5.2 40% 60% 0% Customer loyalty 5.2 40% 60% 0% Employee loyalty 5.1 30% 70% 0% Direct distribution channels 5.0 50% 40% 10% Financial assets 5.0 40% 60% 0% Brand equity 4.9 30% 70% 0% Corporate social responsibility practices (CSR) 4.9 20% 80% 0% Organisational culture 4.8 40% 50% 10% Basic environmental practices 4.5 30% 70% 0% Organisational structure 4.5 30% 70% 0% Advanced environmental 4.3 10% 80% 10% practices The results from round 1 presented the inputs for round 2. A total of 16 drivers of performance with the highest average score from round 1 and additional 4 drivers that were suggested from the panel of experts were included in the questionnaire. The results from the round 2 showed that the drivers with higher average scores were almost the same as in round 1. They included: location (6.4), market orientation (6.4) and customer satisfaction (6.3). Moreover, 80% of the panel experts also rated service quality (6.2), business processes (6.0) and employee competencies (6.0) as important or highly important performance drivers. Employee competencies were in round 2 evaluated higher than in round 1; they also outweighed the importance of management philosophy and competencies. Additionally proposed drivers in the phase 1 of this research: product development, cooperation and flexibility were all rated with an average score of 6.0 or higher (those are marked bold in table 4 and 5). Round 2 of Delphi yielded 20 performance drivers presented in Table 4. Table 4: Delphi round 2 results % of respondents % of respondents % of respondents Mean values that rated the that rated the that rated the Indicator importance as importance as importance as low or high moderate or neutral insignificant (scores (scores 6 or 7) (scores 5, 4 or 3) 2, 1 or 0) Location 6.4 90% 10% 0% Market orientation 6.4 90% 10% 0% Product development 6.3 90% 10% 0% Customer satisfaction 6.3 80% 20% 0% Service quality 6.2 80% 20% 0% Flexibility 6.1 80% 20% 0% Business processes 6.0 80% 20% 0% Employee competencies 6.0 70% 30% 0% Cooperation between 6.0 70% 30% 0% tourism providers on and between destinations Management philosophy 5.9 80% 20% 0% Managerial competencies 5.9 70% 30% 0% HRM practices 5.8 70% 30% 0% Hotel facilities 5.8 60% 40% 0% Employee attitudes toward 5.8 60% 40% 0% work Information technology (IT) 5.7 60% 40% 0% Investment management 5.6 70% 30% 0% Knowledge sharing 5.6 60% 40% 0% Relationship with 5.5 60% 40% 0% commercial and other partners Social capital 5.5 50% 50% 0% Team culture 5.4 50% 50% 0% Source: own research. The process was repeated in round 3. The questionnaire was sent to the panel of experts for another round of evaluation. They were once again asked to evaluate the importance of each driver of hotel performance. In round 3, only location received higher scores (average score of 6.5) and became the most important driver of hotel performance according to the panel of experts. The evaluation showed that the final list was reduced to feature only nine major drivers of hotel performance. Those drivers consistently received the highest scores from the expert panel. They include: location, market orientation, customer satisfaction, product development, service quality, flexibility, business processes, employee competencies, and cooperation. The results from round 3 of expert evaluation are presented in Table 5. Table 5: Final Delphi results - round 3 Indicator Mean Values % of respondents that rated the importance as high (scores 6 or 7) % of respondents that rated the importance as moderate or neutral (scores 5, 4 or 3) % of respondents that rated the importance as low or insignificant (scores 2, 1 or 0) Location 6.5 90% 10% 0% Market orientation 6.4 90% 10% 0% Customer satisfaction 6.3 80% 20% 0% Product development 6.3 90% 10% 0% Service quality 6.2 80% 20% 0% Flexibility 6.1 80% 20% 0% Business processes 6.0 80% 20% 0% Employee competencies 6.0 70% 30% 0% Cooperation between tourism providers on and between destinations 6.0 70% 30% 0% The results were compared with the literature analysis and the Delphi group results shown in Figure 1. The aim was to understand the following: which drivers are examined in relevant literature, but experts do not believe that they are important?; which drivers are recognised as important by experts, but are not getting sufficient attention in relevant literature?; and which drivers are recognised by experts and literature as crucial for hotel success? The results are presented in Figure 1. As visible in section III, the following drives received considerable interest in the literature: HRM practices, brand equity, hotel facilities, environmental practices, organisational culture, and ICT. However, the panel of experts did not recognise those as crucial drivers of hotel success. On the contrary - the panel of experts defined location, firm flexibility, product development and cooperation between tourism providers as crucial drivers of success - as seen in section I. Literature so far did failed to show much interest in those drivers of hotel success. Finally, market orientation, customer satisfaction, service quality and business processes were recognised by both literature and experts as important drivers of hotel success (section II). Figurel: Drivers ofhotel performancethrough literature and the Delphi study 4. DISCUSSION AND CONCLUSION Hotel performance attracted substantial research attention in the past three decades. Mostof the academic workfocusedon empirical investigation connectingsustainable comp ctitrneaclvantggerwrthfinancirl g ^rix^e^emceiae sa vntele. ^udiesst faemainlh Soc^u^cdisn^eegii^i^l^ind ic^te^i^^ft^e^^^eCs^^ce^t^Sic^i^j^en^scl^ion^i end humancarital) sect ^hd^iru^laii^n egi^^e^^^gardc^^^iijt^gcv^ircCs^vgict eomehownerlccted tancible assets aiciSc^^])vb^]^iti^esavtl)Ci(^ti'^^^i^ttt^nh^C^ti gerdormanee.Tungibleascets wcrecnnregers dns a prseevuieitecnd necessaryrnpnt ter overatisns and ids eomrection witdgetabrmonce wasCsegtcd ie aVviousj hi tvm,cavgVciiticrwereneglvcted ^cauie tUrycredif]ccvlt to measure and operationalise in empirical studies. Empirical research in many cases focused on specific drivers of hotel performance and investigated their impact on performance. Most of the studies did yield unexpected results. Tested drivers - sustainable competitive advantages - were translated into financial success of a firm. Most of the studies did not ask the question: which are the most relevant drivers of hotel performance? To this end, this paper is dealing with this question. Qualitative research based on the Delphi method was performed on a panel of 10 experts in the hospitality industry. They evaluated 30 drivers of hotel performance that had been recognised in hospitality research. Three rounds of evaluation reduced the list to nine most important drivers of performance in the hospitality industry by the opinion of the expert panel. The main drivers are: location, customer satisfaction, service quality, employee competencies, business processes, product development, cooperation, flexibility, and market orientation. Of those nine drivers, one is considered as tangible, three are intangible assets and five are capabilities. This is a quite different structure than the structure of the inputs for this research. Among 31 driver of performance, only three were tangible assets, 23 were intangible assets and five of them were capabilities. Theoretical implication of this paper is in indicating the importance of capabilities and tangible assets as performance drivers in hotel industry. Literature was not critically assessing those performance drivers so far. Our guidelines for future research in this area are proposed in section fife of this paper. Managerial implication of this paper is focused, structured and clear communication of relevant performance drivers in hospitality industry. Drivers indicated in this research are areas in which future investments in financial and human capital are needed for sustainable growth and prosperity of the industry. Limitations of this paper are that we have possibly missed some of the research papers in the field. We have studied available paper in the WoS database but body of knowledge is growing and we may overlook some of the work. In terms of method used the main limitation is the subjective evaluation of the respondents and possible impacts of the panel on respondent opinion. Delphi is also narrowly applicable to few specific setting, it is very lengthy and complex to conduct and results off course cannot be generalized. 5. FUTURE RESEARCH POTENTIAL This paper sets the goals for future empirical research in understanding hotel performance. Crossing the literature with hospitality experts' opinions enabled to define potential gaps, which represent a future field of research and operationalisation. Further quantitative research using those concepts is needed in order to generalize the results and further contribute to the field of knowledge in this area. In addition, it would lead to a better understanding of hotel performance in emerging and developed hospitality markets. General management theory is emphasizing the importance of capabilities as prominent performance drivers. The lack of empirical studies investigating the relation between capabilities and performance is evident. However, the results of this research show that hospitality experts believe that capabilities are potential sources of sustainable competitive advantages. Specifically, the panel of experts recognised dynamic capabilities as quite important. Those include a firm's ability to adjust to ongoing changes in the external environment. Adaptability of organisations as systems, their employees and managers is highly relevant in today's ever-changing world. As a result, dynamic capabilities constitute an important research area that should encourage a plethora of research in the future. These research results and general overview of relevant literature underline the need for a shift in research efforts. Research literature improved the understanding of intangible assets and their role in driving the performance in hospitality industry. However, there is still limited information available about capabilities and their potential role as sustainable competitive advantages of a hotel. The development of measurement instruments enabling the evaluation of capabilities and the magnitude of their impact on performance is a potential future avenue in hospitality research. This paper indicated five potential areas of future research including business processes, product development, cooperation, flexibility and market orientation. This paper also compare research efforts delivered in empirical papers measuring drivers of hotel performance with expert opinion on the importance of those drivers. This research identified a gap between the focus in literature and expert evaluations. Experts clearly emphasised the role of tangible assets in hotel performance. The hotel industry is capital intensive and investments in tourism infrastructure are seen as extremely important. The results may be driven by the current situation in the hotel sector in Slovenia and Croatia. The hospitality industry in both countries requires a significant investment cycle to improve its competitiveness on the global market. Furthermore, the experts argued that clear product development is a very important driver of hotel performance. So far, empirical research was not focusing on the product development and its impact on hotel performance. This is also one of potential lines of research indicating the operationalisation of the product development as a driver of success. Business processes related to the revolution in information communication are also recognised as highly important. Qualitative research indicated that it is necessary to address them as a business processes in relation to the customer - all digital communication, internal business processes between hotel employees, and possibly technological solutions that can improve the available tourism products. The panel of experts strongly emphasised market orientation and flexibility as important drivers of performance. Further operationalisation of those drivers is necessary in order to test them empirically. REFERENCES Ahmad, R., Solnet, D. & Scott, N. (2010). Human resource practices system differentiation: A hotel industry study. Journal of Hospitality and Tourism Management, 17(1), 72-82. Al-Refaie, A. (2015). Effects of human resource management on hotel performance using structural equation modeling. Computers in Human Behavior, 43, 293-303. Amit, R. & Schoemaker, P. J. (1993). Strategic assets and organizational rent. Strategic management journal, 14(1), 33-46. Asree, S., Zain, M. & Rizal Razalli, M. (2010). Influence of leadership competency and organizational culture on responsiveness and performance of firms. International Journal of Contemporary Hospitality Management, 22(4), 500-516. Assaf, A. & Cvelbar, K. L. (2011). Privatization, market competition, international attractiveness, management tenure and hotel performance: Evidence from Slovenia. International Journal of Hospitality Management, 30(2), 391-397. Assaf, A. G., Josiassen, A. & Cvelbar, L. K. (2012). Does triple bottom line reporting improve hotel performance? International Journal of Hospitality Management, 31(2), 596-600. Assaf, A. G., Josiassen, A., Cvelbar, L. K. & Woo, L. (2015). The effects of customer voice on hotel performance. International Journal of Hospitality Management, 44, 77-83. Assaf, A. G. & Magnini, V. (2012). Accounting for customer satisfaction in measuring hotel efficiency: Evidence from the US hotel industry. International Journal of Hospitality Management, 31(3), 642-647. Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of management, 17(1), 99-120. Barros, C. P. & Dieke, P. U. (2008). Technical efficiency of African hotels. International Journal of Hospitality Management, 27(3), 438-447. Benavides-Velasco, C. A., Quintana-García, C. & Marchante-Lara, M. (2014). Total quality management, corporate social responsibility and performance in the hotel industry. International Journal of Hospitality Management, 41, 77-87. Božič, V. & Kneževic Cvelbar, L. (2016). Resources and capabilities driving performance in the hotel industry. Tourism and Hospitality Management, 22(2), 225-246. Chand, M. (2010). The impact of HRM practices on service quality, customer satisfaction and performance in the Indian hotel industry. The International Journal of Human Resource Management, 21(4), 551-566. Chand, M. & Katou, A. A. (2007). The impact of HRM practices on organisational performance in the Indian hotel industry. Employee Relations, 29(6), 576-594. Chi, C. G. & Gursoy, D. (2009). Employee satisfaction, customer satisfaction, and financial performance: An empirical examination. International Journal of Hospitality Management, 28(2), 245-253. Cho, S., Woods, R. H., Jang, S. S. & Erdem, M. (2006). Measuring the impact of human resource management practices on hospitality firms' performances. International Journal of Hospitality Management, 25(2), 262-277. Chu, R. K. & Choi, T. (2000). An importance-performance analysis of hotel selection factors in the Hong Kong hotel industry: a comparison of business and leisure travellers. Tourism management, 21(4), 363-377. Chung-Herrera, B. G., Enz, C. A. & Lankau, M. J. (2003). Grooming future hospitality leaders: A competencies model. The Cornell Hotel and Restaurant Administration Quarterly, 44(3), 17-25. Claver-Cortés, E., Pereira-Moliner, J., José Tarí, J. & Molina-Azorín, J. F. (2008). TQM, managerial factors and performance in the Spanish hotel industry. Industrial Management & Data Systems, 108(2), 228-244. Claver, E., José Tarí, J. & Pereira, J. (2006). Does quality impact on hotel performance?. International Journal of Contemporary Hospitality Management, 18(4), 350-358. Cohen, J. F. & Olsen, K. (2013). The impacts of complementary information technology resources on the serviceprofit chain and competitive performance of South African hospitality firms. International journal of hospitality management, 34, 245-254. Collis, D. J. (1994). Research note: how valuable are organizational capabilities?. Strategic management journal, 15(S1), 143-152. Dai, W. D., Mao, Z. E., Zhao, X. R. & Mattila, A. S. (2015). How does social capital influence the hospitality firm's financial performance? The moderating role of entrepreneurial activities. International Journal of Hospitality Management, 51, 42-55. De Grosbois, D. (2012). Corporate social responsibility reporting by the global hotel industry: Commitment, initiatives and performance. International Journal of Hospitality Management, 31(3), 896-905. Fu, H., Ye, B. H. & Law, R. (2014). You do well and I do well? The behavioral consequences of corporate social responsibility. International Journal of Hospitality Management, 40, 62-70. Garay, L. & Font, X. (2012). Doing good to do well? Corporate social responsibility reasons, practices and impacts in small and medium accommodation enterprises. International Journal of Hospitality Management, 31(2), 329-337. http://dx.doi.org/10.1016/j.ijhm.2011.04.013 Gray, B. J., Matear, S. M. & Matheson, P. K. (2000). Improving the performance of hospitality firms. International Journal of Contemporary Hospitality Management, 12(3), 149-155. Hall, R. (1992). The strategic analysis of intangible resources. Strategic management journal, 13(2), 135-144. Hernández-Perlines, F. (2016). Entrepreneurial orientation in hotel industry: multi-group analysis of quality certification. Journal of Business Research, 69(10), 4714-4724. Harrington, D. & Akehurst, G. (1996). Service quality and business performance in the UK hotel industry. International Journal of Hospitality Management, 15(3), 283-298. Hoque, K. (1999). Human resource management and performance in the UK hotel industry. British Journal of Industrial Relations, 37(3), 419-443. Hu, M. L. M., Horng, J. S. & Sun, Y. H. C. (2009). Hospitality teams: Knowledge sharing and service innovation performance. Tourism management, 30(1), 41-50. Hussain, K., Konar, R. & Ali, F. (2016). Measuring Service Innovation Performance through Team Culture and Knowledge Sharing Behaviour in Hotel Services: A PLS Approach. Procedia-Social and Behavioral Sciences, 224, 35-43. Jarboui, S., Guetat, H. & Boujelbène, Y. (2015). Evaluation of hotels performance and corporate governance mechanisms: Empirical evidence from the Tunisian context. Journal of Hospitality and Tourism Management, 25, 30-37. Jeou-Shyan, H., Hsuan, H., Chih-Hsing, L., Lin, L. & Chang-Yen, T. (2011). Competency analysis of top managers in the Taiwanese hotel industry. International Journal of Hospitality Management, 30(4), 10441054. Jogaratnam, G. & Ching-Yick Tse, E. (2006). Entrepreneurial orientation and the structuring of organizations: performance evidence from the Asian hotel industry. International Journal of Contemporary Hospitality Management, 18(6), 454-468. Josiassen, A., Assaf, A. G. & Cvelbar, L. K. (2014). CRM and the bottom line: Do all CRM dimensions affect firm performance?. International Journal of Hospitality Management, 36, 130-136. http://dx.doi.org/10.1016/). ijhm.2013.08.005 Kandampully, J. & Suhartanto, D. (2000). Customer loyalty in the hotel industry: the role of customer satisfaction and image. International journal of contemporary hospitality management, 12(6), 346-351. Kang, K. H., Lee, S. & Huh, C. (2010). Impacts of positive and negative corporate social responsibility activities on company performance in the hospitality industry. International Journal of Hospitality Management, 29(1), 72-82. Kay, C. & Russette, J. (2000). Hospitality-management competencies: Identifying managers' essential skills. The Cornell Hotel and Restaurant Administration Quarterly, 41(2), 52-63. Kemp, S. & Dwyer, L. (2001). An examination of organisational culture—the Regent Hotel, Sydney. International journal of hospitality management, 20(1), 77-93. Kim, H. B. & Kim, W. G. (2005). The relationship between brand equity and firms' performance in luxury hotels and chain restaurants. Tourism management, 26(4), 549-560. Kim, T. T., Kim, W. G., Park, S. S. S., Lee, G. & Jee, B. (2012). Intellectual Capital and Business Performance: What Structural Relationships Do They Have in Upper-Upscale Hotels?. International Journal of Tourism Research, 14(4), 391-408. Kim, M., Vogt, C. A. & Knutson, B. J. (2015). Relationships among customer satisfaction, delight, and loyalty in the hospitality industry. Journal of Hospitality & Tourism Research, 39(2), 170-197. Kim, W. G. & Brymer, R. A. (2011). The effects of ethical leadership on manager job satisfaction, commitment, behavioral outcomes, and firm performance. International Journal of Hospitality Management, 30(4), 1020-1026. Kim, W. G., Cho, M. & Brymer, R. A. (2013). Determinants affecting comprehensive property-level hotel performance: The moderating role of hotel type. International Journal of Hospitality Management, 34, 404-412. Knezevic Cvelbar, L. & Mihalic, T. (2007). Ownership structure as a corporate governance mechanism in Slovenian hotels. Privredna kretanja i ekonomskapolitika, 17(112), 26-51. Kucukusta, D., Mak, A. & Chan, X. (2013). Corporate social responsibility practices in four and five-star hotels: Perspectives from Hong Kong visitors. International Journal of Hospitality Management, 34, 19-30. Lado-Sestayo, R., Otero-González, L., Vivel-Búa, M. & Martorell-Cunill, O. (2016). Impact of location on profitability in the Spanish hotel sector. Tourism Management, 52, 405-415. Leonidou, L. C., Leonidou, C. N., Fotiadis, T. A. & Zeriti, A. (2013). Resources and capabilities as drivers of hotel environmental marketing strategy: Implications for competitive advantage and performance. Tourism Management, 35, 94-110. Matear, S., Osborne, P., Garrett, T. & Gray, B. J. (2002). How does market orientation contribute to service firm performance? An examination of alternative mechanisms. European Journal of Marketing, 36(9/10), 1058-1075. McKenna, H. P. (1994). The Delphi technique: a worthwhile research approach for nursing?. Journal of advanced nursing, 19(6), 1221-1225. Melián-González, S. & Bulchand-Gidumal, J. (2016). A model that connects information technology and hotel performance. Tourism Management, 53, 30-37. Mihalic, T. & Buhalis, D. (2013). ICT as a new competitive advantage factor-Case of small transitional hotel sector. Economic and Business Review for Central and South-Eastern Europe, 15(1), 33. Mihalic, T., Garbin Pranicevic, D. & Arneric, J. (2015). The changing role of ICT competitiveness: the case of the Slovenian hotel sector. Ekonomska istrazivanja, 28(1), 367-383. Molina-Azorín, J. F., Claver-Cortés, E., Pereira-Moliner, J. & Tarí, J. J. (2009). Environmental practices and firm performance: an empirical analysis in the Spanish hotel industry. Journal of Cleaner Production, 17(5), 516-524. Molina-Azorín, J. F., Tarí, J. J., Pereira-Moliner, J., López-Gamero, M. D. & Pertusa-Ortega, E. M. (2015). The effects of quality and environmental management on competitive advantage: A mixed methods study in the hotel industry. Tourism Management, 50, 41-54. Naseem, A., Sheikh, S. E. & Malik, K. P. (2011). Impact of employee satisfaction on success of organization: Relation between customer experience and employee satisfaction. International journal of multidisciplinary sciences and engineering, 2(5), 41-46. Nieves, J., Quintana, A. & Osorio, J. (2014). Knowledge-based resources and innovation in the hotel industry. International Journal of Hospitality Management, 38, 65-73. Oltean, F. D., Gabor, M. R. & Contiu, L. C. (2014). Relation between Information Technology and Performance: An Empirical Study Concerning the Hotel Industry in Mures County. Procedia Economics and Finance, 15, 1535-1542. O'Neill, J. W. & Carlbäck, M. (2011). Do brands matter? A comparison of branded and independent hotels' performance during a full economic cycle. International Journal of Hospitality Management, 30(3), 515-521. 0gaard, T., Marnburg, E. & Larsen, S. (2008). Perceptions of organizational structure in the hospitality industry: Consequences for commitment, job satisfaction and perceived performance. Tourism Management, 29(4), 661671. Pereira-Moliner, J., Claver-Cortés, E., Molina-Azorín, J. F. & Tarí, J. J. (2012). Quality management, environmental management and firm performance: direct and mediating effects in the hotel industry. Journal of Cleaner Production, 37, 82-92. Pereira-Moliner, J., Font, X., Tarí, J. J., Molina-Azorin, J. F., Lopez-Gamero, M. D. & Pertusa-Ortega, E. M. (2015). The Holy Grail: Environmental management, competitive advantage and business performance in the Spanish hotel industry. International Journal of Contemporary Hospitality Management, 27(5), 714-738. Prasad, K. & Dev, C. S. (2000). Managing hotel brand equity: A customer-centric framework for assessing performance. The Cornell Hotel and Restaurant Administration Quarterly, 41(3), 224-31. Rudez, H. N. & Mihalic, T. (2007). Intellectual capital in the hotel industry: A case study from Slovenia. International Journal of Hospitality Management, 26(1), 188-199. Ruzic, M. D. (2015). Direct and indirect contribution of HRM practice to hotel company performance. International Journal of Hospitality Management, 49, 56-65. Sharpley, R. & Forster, G. (2003). The implications of hotel employee attitudes for the development of quality tourism: the case of Cyprus. Tourism management, 24(6), 687-697. Sirirak, S., Islam, N. & Ba Khang, D. (2011). Does ICT adoption enhance hotel performance?. Journal of Hospitality and Tourism Technology, 2(1), 34-49. Sorensen, J. B. (2002). The strength of corporate culture and the reliability of firm performance. Administrative science quarterly, 47(1), 70-91. Srivastava, A., Bartol, K. M. & Locke, E. A. (2006). Empowering leadership in management teams: Effects on knowledge sharing, efficacy, and performance. Academy of management journal, 49(6), 1239-1251. Sun, K. A. & Kim, D. Y. (2013). Does customer satisfaction increase firm performance? An application of American Customer Satisfaction Index (ACSI). International Journal of Hospitality Management, 35, 68-77. Tarí, J. J., Claver-Cortés, E., Pereira-Moliner, J. & Molina-Azorín, J. F. (2010). Levels of quality and environmental management in the hotel industry: Their joint influence on firm performance. International Journal of Hospitality Management, 29(3), 500-510. Tavitiyaman, P., Qiu Zhang, H. & Qu, H. (2012). The effect of competitive strategies and organizational structure on hotel performance. International Journal of Contemporary Hospitality Management, 24(1), 140-159. Terry Kim, T., Lee, G., Paek, S. & Lee, S. (2013). Social capital, knowledge sharing and organizational performance: what structural relationship do they have in hotels?. International Journal of Contemporary Hospitality Management, 25(5), 683-704. Vega-Vázquez, M., Cossío-Silva, F. J. & Revilla-Camacho, M. Á. (2016). Entrepreneurial orientation-hotel performance: Has market orientation anything to say?. Journal of Business Research, 69(11), 5089-5094. Wang, C. H., Chen, K. Y. & Chen, S. C. (2012). Total quality management, market orientation and hotel performance: The moderating effects of external environmental factors. International Journal of Hospitality Management, 3J(1), 119-129. Wilkins, H., Merrilees, B. & Herington, C. (2007). Towards an understanding of total service quality in hotels. International Journal of Hospitality Management, 26(4), 840-853. Wu, C. M. & Chen, T. J. (2015). Psychological contract fulfillment in the hotel workplace: Empowering leadership, knowledge exchange, and service performance. International Journal of Hospitality Management, 48, 27-38. Xiao, Q., O'Neill, J. W. & Mattila, A. S. (2012). The role of hotel owners: the influence of corporate strategies on hotel performance. International Journal of Contemporary Hospitality Management, 24(1), 122-139. Zoghbi-Manrique-de-Lara, P. & Ting-Ding, J. M. (2016). The influence of corporate culture and workplace relationship quality on the outsourcing success in hotel firms. International Journal of Hospitality Management, 56, 66-77. THE INTERPLAY AMONG PROSOCIAL MOTIVATION, CULTURAL TIGHTNESS, AND UNCERTAINTY AVOIDANCE IN PREDICTING KNOWLEDGE HIDING Katja Babič1 Matej Černe2 Received: April 10, 2017 Accepted: May 18, 2018 Miha Škerlavaj3 Pengcheng Zhang4 ABSTRACT: In this paper, we examine the socio-cultural aspects of knowledge hiding. Specifically, we aim to deepen the understanding regarding the role that national cultural dimensions and motivation play in stimulating or preventing knowledge hiding. We investigate a three-way interaction among prosocial motivation, cultural tightness, and uncertainty avoidance to explain knowledge hiding in organizations. Our field studies involved working professionals from Slovenia (n = 123) and China (n = 253). Results show that the highest level of knowledge hiding happens when employees are met with a combination of a low level of prosocial motivation, a low level of cultural tightness, and a low level of uncertainty avoidance. The highest levels of knowledge hiding thus occur when employees are not motivated by the welfare of others, are inclined to take the risk and simultaneously perceive that deviation from cultural norms will not be sanctioned. We discuss the contributions and implications of our two studies for the fields of knowledge hiding and cross-cultural organizational behavior. Keywords: prosocial motivation, cultural tightness-looseness, uncertainty avoidance, knowledge hiding JEL: D83 DOI: 10.15458/85451.71 INTRODUCTION Knowledge hiding is an emerging construct that is quickly gaining ground in contemporary research. This construct presents treatment as an interpersonal phenomenon (Connelly et al., 2012) and it has been recognized as a pressing matter in the workplace (Cui, Park. & Paik, 2016; Cerne, Nerstad, Dysvik. & Skerlavaj, 2014). Interpersonal distrust and poor relationships in the workplace positively influence knowledge hiding. Moreover, knowledge hiding harms interpersonal relationships (Connelly et al., 2014), increases 1 Corresponding author, University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: katja. babic@ef.uni-lj.si> 2 University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: matej.cerne@ef.uni-lj.si> 3 University of Ljubljana, Faculty of Economics, Ljubljana, Slovenia, e-mail: miha.skerlavaj@ef.uni-lj.si> 4 University of Science & Technology, China distrust (Černe et al., 2014; Grant & Sumanth, 2009; Oye, Salleh. & Noorminshah, 2011), diminishes creativity (Černe et al., 2014) and innovative work behavior (Černe et al., 2017), causes lower level of motivation (Oye et al., 2011), and creativity (Connelly et al., 2012; Černe et al., 2014). Based on the previous research, we learned that prosocial motivation is a significant predictor of knowledge hiding (Černe et al., 2014; Černe, Babič, Connelly. & Škerlavaj, 2015). However, boundary conditions of the relationship between prosocial motivation and knowledge hiding (Connelly et al., 2012; Grant, 2007) remain largely unexplored. It is important to be aware of the outputs and consequences that this relationship could bring out. Prosocial motivation protects and promotes the well-being of others, so it should play an important role in predicting knowledge hiding (Batson, 1987). It refers to a desire to help others (Ardichvili, Page. & Wentling, 2003; Batson, 2010; Miller, 1994). Based on that knowledge, we focus on how prosocial motivation (Bolino & Grant, 2016; Grant, 2007), in connection with national cultural dimensions of tightness-looseness and uncertainty avoidance (Gelfand, Erez. & Aycan, 2007; Tsui, Nifadkar. & Ou, 2007), influences knowledge hiding. Even though in our research we combined two compatible cultural dimensions, there are some important differences between them. With cultural tightness, we measure the strength or intensity, and with uncertainty avoidance, we measure the content of national culture. The difference is also that uncertainty avoidance is a cultural dimension that deals with a society's tolerance for uncertainty and ambiguity (Hofstede, 1983a). On the other hand, cultural tightness is a construct that presents the strength of social norms and degree of sanctioning within societies (Gelfand, Nishii. & Raver, 2006). Cultural tightness is a broader construct than uncertainty avoidance; although cultures could be tight, not all tight cultures are uncertainty-avoidance oriented. Today, organizations work globally and it is crucial to learn about the cultural context to reduce misunderstanding and inappropriate behavior in advance (Johnson, Lenartowicz. & Apud, 2006). Organizations need to be aware that each culture has its own norms, values, and official rules that influence work behavior (Aktas, Gelfand. & Hanges, 2015; Chua, Roth. & Lemoine, 2015; Gelfand et al., 2007; Gelfand, Nishii. & Raver, 2006). The extant research shows that work behavior is influenced by various cultural norms (Chua et al., 2015; Pan. & Zhang, 2014). Cultural tightness and uncertainty avoidance have already been demonstrated to have a significant influence on organizational behavior (Fischer, Ferreira, Leal, Redford. & Harb, 2005; Gelfand et al., 2007; Liu, Jiang, Shalley, Keem. & Zhou, 2016). Uncertainty avoidance, one of the core Hofstede's cultural dimensions, is defined as the extent to which the members of a culture feel threatened by ambiguous or unknown situations, and holds important research opportunities (Doney, Cannon. & Mullen, 1998; Litvin, Crotts. & Hefner, 2004). We included the construct of uncertainty avoidance in our research because it is correlated with the construct of cultural tightness (Chua et al., 2015). Cultural tightness essentially represents strengths of cultural norms. Cultural tightness is related with uncertainty avoidance, because both constructs describe strong rules, norms, and sanctions for inappropriate behavior (Chan, 1996; Gelfand et al., 2006; Gelfand, Lim. & Raver, 2004; Toh & Leonardelli, 2013; Triandis, 1977; Triandis, 1998; Triandis, 2000). Based on previous research (Cerne et al., 2015), we have learned that implication of many rules and sanctions could prevent knowledge hiding. In fact, it was found that a three-way interaction term between low prosocial motivation, low cultural tightness, and low collectivism caused the occurrences of knowledge hiding at higher levels - more often, frequently and with a higher chance (Cerne et al., 2015). Therefore, a connection between the constructs of uncertainty avoidance and cultural tightness exists based on their common characteristics (i.e., with regard to their settled rules and their lack of deviation from those rules). In this study, we propose that knowledge hiding is conditioned by prosocial motivation, cultural tightness, and with uncertainty avoidance. The three-way interaction term in a combination of the researched constructs showed the importance of the effect of socio-cultural aspects on knowledge hiding. Uncertainty avoidance means avoiding criticism, negative feedback, and conflict. Avoiding challenges carries the risk of errors and failure. When we are facing low uncertainty avoidance (Shane, 1995; Wennekers, Thurik, van Stel. & Noorderhaven, 2007) this increases the negative way of behavior (Lewicki, McAllister. & Bies, 1998; Tsai, 2002) within an organization. With our study, we empirically demonstrate that managers need to be aware of the occurrences of the combination of low levels of prosocial motivation, low cultural tightness, and low levels of uncertainty avoidance, which leads to knowledge hiding at the highest level. Moreover, managers need to pay special attention, if knowledge hiding occurs among their employees. They should be aware when a combination of three-way interaction term occurs within their organization. This paper intends to contribute to the field of knowledge hiding. We studied two samples of working professionals, from two distinctly different settings - China and Slovenia, to increase the generalizability of findings. To predict knowledge hiding, we considered the type of culture using uncertainty avoidance and cultural tightness. Second, this study intends to contribute to cross-cultural organizational behavior research. We demonstrated the importance of combining various socio-cultural aspects using a three-way interaction term, and we connected cultural tightness (i.e., the strength) to uncertainty avoidance (i.e., the content). Previously, this combined approach, using two theoretical frameworks focused on national cultures, has been underexplored (Gelfand et al., 2007; Tsui et al., 2007). Furthermore, in our research, we focused on variations within cultural dimensions from the individual-level assessments at the micro level (Brewer & Venaik, 2014; Taras, Steel. & Kirkman, 2016; Tsui et al., 2007). Variations in cultural values were found to mainly occur within countries and much less across, as a meta-analysis showed. Taras et al. (2016) gathered all the Hofstede-typology based studies to demonstrate that 80% of variance occurs on between-person level, and only 20% among different countries. When studying culture and values, individual perceptions are thus a stronger placeholder for cultural dimensions than countries are. THEORY AND HYPOTHESES Knowledge hiding Knowledge hiding is defined as an intentional attempt to conceal or withhold information that others have requested (Connelly et al., 2012). It(Connelly, 2012 #713) occurs when an employee receives a request for knowledge, but actively conceals or withholds that knowledge. This definition specifically excludes cases in which employees fail to share knowledge due to mistakes, accidents, or ignorance, as well as cases when they simply do not have the requested knowledge. Prior research describes three facets of knowledge hiding, which are: playing dumb (i.e., an individual pretends not to know the relevant information), evasive hiding, (i.e., individual pretends that the information will be forthcoming even as he or she intends to conceal it) and rationalized hiding (individual provides an accurate explanation to explain why the information will not be forthcoming) (Connelly et al., 2012). Knowledge hiding, not simply the opposite of knowledge sharing (Connelly et al., 2012), includes intentional refusals to share in response to requests. Up to this date, knowledge hiding remains a novel and unexplored phenomenon. An example of knowledge hiding is when an employee receives a request for knowledge and this employee engages in activities designed to conceal the requested knowledge (Connelly et al., 2012). In other words, employees who possess knowledge decline to share it without the expectation of reciprocity (Davenport & Prusak, 1998). Prosocial motivation and knowledge hiding We included the construct of prosocial motivation as a crucial social predictor of knowledge hiding. Prosocial motivation represents a desire to expand effort based on a concern for helping or contributing to other people (Grant, 2007). Prosocially motivated employees are more likely to foster cooperative behavior with their colleagues in teams (Bolino & Grant, 2016; Hu & Liden, 2015). These employees have a high chance of being treated like good and trusted employees. However, these employees could also be exploited by free-riders (Lenway & Rehbein, 1991). Prosocially motivated employees appreciate values, norms, and a positive workplace climate (Ardila, Gouveia. & Diogenes de Medeiros, 2012; Schwartz, 2007; Skerlavaj, Connelly, Cerne. & Dysvik, In press). People without social empathy engage in knowledge hiding (Cui, Park. & Paik, 2016). In previous studies, researchers have indicated that prosocial motivation is potentially negatively related to knowledge hiding (Cerne et al., 2015). The research of combination of two constructs prosocial motivation and knowledge hiding is novel. Actually, Cerne et al. (2015) were the first ones who were interested in examining this relationship, and supported the fact that prosocial motivation negatively influences knowledge hiding (Cerne et al., 2015). Previous research of this link (Cerne et al., 2015) was made based on two studies, but with a lower sample (Study 1: n = 115; Study 2: n = 313). The present study on a larger sample, with more participants from different cultures, replicates and strengthens previous research findings. Therefore, we used the following hypothesis as a basis for further hypothesized interactions: Hypothesis 1: Prosocial motivation is negatively related to knowledge hiding. Cross-cultural knowledge hiding In our research, we have focused on the study of cross-cultural aspects in knowledge hiding, especially on how culture could influence behavior within an organization. We have used respondents from two countries and measured their personal cultural tightness and uncertainty avoidance. As many companies operate globally (Williams, Han. & Qualls, 1998), cross-cultural research is gaining important value (Berry, 2002; Hui & Triandis, 1986). Taras, Steel, and Kirkman (2016) found that over 80% of the variations in cultural values were found within countries and less than 20% were found between countries. Therefore, it is important to study each culture separately, to learn which types of behavior and which personal characteristics prevail within each culture (Taras et al., 2016). Nevertheless, knowledge hiding among employees remains a problem. When employees experience knowledge hiding behavior from their coworkers (Connelly et al., 2012), this leads to a lack of trust and to a bad working environment. Regardless of cultural characteristics, leaders are those who should support knowledge-sharing environment within an organization (Connelly & Kelloway, 2003; De Vries, Bakker-Pieper. & Oostenveld, 2010). In other words, irrespective of the national culture in which an organization exists, leaders and followers should create their own organizational culture (Al-Alawi, Al-Marzooqi. & Mohammed, 2007; Davenport & Prusak, 1998; Oye et al., 2011; Suppiah & Singh Sandhu, 2011). Based on this, cultural aspects are tightly related to organizational behavior. Researchers proved that there is an influence of cultural individualism-collectivism, self-construal, and individual values on communication styles across cultures (Gudykunst, Matsumoto, Ting-Toomey, Nishida, Kim. & Heyman, 1996). In addition, it depends on the cultural aspects of organizational behavior whether individuals will be prepared to share or instead hide their knowledge (Bock, Zmud, Kim. & Lee, 2005). Two-way interaction: Cultural tightness-looseness and uncertainty avoidance predicting knowledge hiding In the following, we describe both researched constructs studied in our two-way interaction term research. Firstly, we present the cultural tightness characteristics and how this could influence knowledge hiding. Secondly, we present uncertainty avoidance characteristic and its influence on knowledge hiding. After this, we combined two researched constructs in a two-way interaction term of cultural tightness and uncertainty avoidance predicting knowledge hiding. Cultural tightness-looseness and knowledge hiding The prevailing norms of a national culture could influence knowledge hiding. Norms are shared beliefs and expectations of what behavior is appropriate within a culture (Cialdini, Kallgren. & Reno, 1991). Culture is an expression of deeper norms and values in a society. It cannot be directly seen and is quite difficult to identify (Trompenaars & Hampden-Turner, 2011). Observing the particularities of a culture means studying that culture on the individual level (Earley & Mosakowski, 2000; McSweeney, 2002; Salk & Brannen, 2000). Due to the fact that cultural differences are found to occur within a country (Taras et al., 2012), this tells us that it should be studied at the individual level. Pelto (1968) was the first to divide cultures into tight and loose societies based on their social norms. Cultural norms, values, and cultural differences influence organizational behavior, especially when individuals work in teams (Cox, Lobel. & McLeod, 1991). Cultural tightness is defined as a construct with strong social norms and sanctions for inappropriate behavior (Gelfand et al., 2006; Toh & Leonardelli, 2013; Triandis, 1977; Triandis & Gelfand, 1998). Social norms and sanctions highly influence the behavior of individuals within organizations (Gelfand et al., 2006). Tight societies have clearly defined norms, values, rules, and require strict discipline (Chua et al., 2015; Gelfand et al., 2006; Shin, Hasse. & Schotter, 2015). Examples of tight cultures are the Japanese, Indian, and Hutterite societies (Pelto, 1968). Chua and colleagues (2015) defined loose cultures as those that have the opposite characteristics of tight cultures. A loose culture lacks discipline, formality, and regimentation, and it has a high tolerance for deviant behavior (Pelto, 1968). Examples of loose cultures are the Finnish and Thai societies (Pelto, 1968). Individuals who possess the requested knowledge have the power, and they decide if they are prepared to share or hide the requested knowledge (Nonaka, 1994). The general characteristics of a culture reveal how people behave overall (Avolio & Bass, 1995; Berry, 1995). Based on the above characteristics, it is expected that cultural tightness could be negatively related to knowledge hiding. Uncertainty avoidance and knowledge hiding Hofstede (1980a; 2011) introduced a model of six dimensions of national culture (i.e., individualism or collectivism, power distance, uncertainty avoidance, masculinity or femininity, long term or short term orientation, and indulgence or restraint) to describe the values of people within particular cultures. These six cultural dimensions (Hofstede, 2011) can be used to describe cultural differences, which prevail in a national culture. According to a literature review, one of six cultural dimensions (i.e., uncertainty avoidance) remains largely unexplored (Conner, Reardon, Miller, Salciuviene. & Auruskeviciene, 2017; Hofstede, 2011; Javidan, House, Dorfman, Hanges. & De Luque, 2006; Johnson, Kulesa, Cho. & Shavitt, 2005; Taras et al., 2012). Countries that exhibit strong uncertainty avoidance are intolerant of inappropriate behavior and ideas. People in cultures with high uncertainty avoidance tend to behave more rationally (Doney, Cannon. & Mullen, 1998; Hofstede, Hofstede. & Minkov 1991; Litvin, Crotts. & Hefner, 2004). The society will try to control the future or just behave in the "let it happen" manner (Hofstede, Hofstede. & Minkov 1991). Hofstede (1983a) described cultures with high uncertainty avoidance as those that strive to have formal and official rules. For these cultures, opposition to rules creates stress. Rules in national culture present feelings of security by helping people to avoid the unfamiliar. Individuals that perceive their cultures as uncertainty avoidant do not want to take risks; they prefer familiar situations (Hofstede, 1983a). Uncertainty avoidance can be understood as a country-level characteristic that affects the appraisal of stressors at the individual level (Debus, Probst, Konig. & Kleinmann, 2012; Hofstede, 2001). It reflects the extent to which ambiguous situations are considered threatening within a society (Hofstede, 2001). Previous studies have shown that uncertainty avoidance is a cultural value that explains the variations in national rates of innovation (Shane, 1993). Additionally, uncertainty avoidance is associated with certainty and predictability (Steers, Meyer. & Sanchez-Runde, 2008). Shane (1993) has also shown that national culture influences activity in development and research. Using similar lenses but different foci, researchers have shown that those employees in cultures that have high levels of uncertainty avoidance feel that everything should be clear and organized in advance (Chua et al., 2015; Gelfand et al., 2007; Hofstede, 1980a; Hofstede, 1980b; Pelto, 1968). Wennekers, Thurik, Van Stel, and Noorderhaven (2007) conducted a study over three years and found important correlations between uncertainty avoidance and some other important economic constructs (i.e., risk, entrepreneurship, GDP per capita). Researchers have also found some good examples of cultures with high levels of uncertainty avoidance: Russia, Korea, Belgium, and Italy (Hofstede, 1993). On the opposite side, there are cultures with low levels of uncertainty avoidance. These cultures have few rules and relatively little structure or stress. Examples of low uncertainty-avoidance countries are India, Indonesia, Singapore, and the United Kingdom (Hofstede, 1993). In this study, we focus on studying the effect of the national cultural dimension of uncertainty avoidance on knowledge hiding. Researchers have shown that, in high uncertainty avoidance countries, employees accept roles and are innovative, rational, and organized (Chua et al., 2015; Gelfand et al., 2007). Based on these characteristics of uncertainty avoidance, we learn that many cultural values can influence individuals' lives. Furthermore, cultural values also influence whether a person may be prepared to share his or her knowledge. Moreover, sharing of knowledge is especially needed in industry, especially for information technology firms, laboratories, newly opened or start-up firms etc. (Hall & Saias, 1980; Waters, 1980). Moreover, it was researched that knowledge sharing leads to higher levels of creativity and innovation (Damanpour & Aravind, 2012). In these cases, employees who do not want to share the requested knowledge would be considered as inappropriate and unwanted. In particular, hiding knowledge from coworkers is unfair toward those who strive for the success of the whole organization. Regarding the mentioned, countries that have high levels of uncertainty avoidance, have specific rules and strict norms with sanctions, and they strive to prevent negative behavior (i.e., knowledge hiding). Combination of cultural tightness and uncertainty avoidance predicting knowledge hiding The construct of cultural tightness-looseness was established in the 1960s; however, it is still an underexplored construct with some open research questions. Until now, researchers have still not focused enough on understanding whether cultural tightness-looseness can have a determining influence on knowledge hiding or not. Moreover, our combined approach—using the interaction of two national cultural dimensions, cultural tightness, and uncertainty avoidance—has not been explored together (Gelfand et al., 2007; Tsui et al., 2007). Both of the researched constructs, cultural tightness, and uncertainty avoidance, share common characteristics about strict rules, norms, and values, as well as strong discipline and sanctions for deviant behavior (Chua et al., 2015; Gelfand et at., 2006). However, there are also differences; as an example, cultural tightness relates to cultural strength or intensity, whereas uncertainty avoidance focuses on cultural content. The main reason that tight cultures need strongly defined rules is the presence of human-made social threats in those nations. Employees who have clearly defined rules and norms (i.e., tightness) do not want to take risks to break up those rules (i.e., uncertainty avoidance). The characteristics of these two constructs are intertwined, and an organization needs to settle them so that employees can follow the written rules. With clearly defined rules and norms, these cultures encourage their populations to behave in an expected way. With strict norms, these cultures can control their populations and coordinate their social actions to ensure survival. For organizations in which rules and sanctions prevail, the sharing of knowledge and information is an encouraging way of behavior. When negative behavior (i.e., knowledge hiding) occurs, it is immediately sanctioned by their leaders or superiors. The combination of uncertainty avoidance and cultural tightness stimulates positive elements of organizational behavior (Hofstede, 1980a). When the organization does not ensure properly designed rules (i.e., when it has low cultural tightness), people might take risks (i.e., low uncertainty avoidance) and could sometimes break the rules without suffering any sanctions; thus, a high level of knowledge hiding could occur. The term "high level" means that this leads to greater occurrences of knowledge hiding. In other words, a high level of knowledge hiding could occur when an organization has a low level of both cultural tightness and uncertainty avoidance. For that reason, we expect that employees with high levels of both uncertainty avoidance and cultural tightness could prevent high levels of knowledge hiding from occurring. Hence, we propose the following hypothesis: Hypothesis 2. Interaction of cultural tightness and uncertainty avoidance is associated with lower levels of knowledge hiding. Three-way interaction: Prosocial motivation, cultural tightness and uncertainty avoidance predicting knowledge hiding In this part, we focus on how the three previously proposed constructs jointly predict the occurrence of knowledge hiding. The combination of cultural tightness and uncertainty avoidance, in connection with prosocial motivation, could prevent knowledge hiding. The suggested way of behavior could support positive behavior. Previous research has shown that prosocial motivation negatively influences knowledge hiding because of the accompanying desire to help others (Cerne et al., 2015). The second and third researched constructs (i.e. cultural tightness and uncertainty avoidance) share some characteristics. Both constructs focus on culture; both support norms and strict rules with sanctions; and both strive to keep everything in order (Aktas et al., 2015; Chan, 1996; Zhang & Zhou, 2014). Furthermore, cultural tightness and uncertainty avoidance support control and require the organization to avoid doubts. Connecting cultural tightness and uncertainty avoidance with prosocial motivation results in an organization's preferred types of behavior. If an organization implements the suggested three-way interaction of prosocial motivation, cultural tightness, and uncertainty avoidance, it could prevent knowledge hiding. Because both of this combination's cultural constructs have strict rules and both appropriately sanction any unexpected or deviant behavior, such as knowledge hiding (Chua et al., 2015). The suggested three-way interactions can help an organization to establish appropriate behavior. More precisely, all three researched constructs describe types of individual behavior that are focused on group benefits, motivation to help others, and caring about the common good (Chen et al., 1998; Kagitcibasi, 1997). Each of the constructs plays a positive role in stimulating beneficial organizational-behavior outcomes (cf. Chua et al., 2015; Gelfand et al., 2006; Gelfand et al., 2007; Grant, 2007). We join all three constructs—prosocial motivation, cultural tightness, and uncertainty avoidance—based on individual perceptions in form of a three-way interaction term. Previous studies of Grant (2007), Gelfand et al. (2006, 2007), and Chua with colleagues (2015) found that each of the researched constructs has a positive influence on organizational behavior. Therefore, this term can be expected to have an even stronger effect on appropriate behavior within the organization. The three-way interaction, supporting positive behavior, could prevent knowledge hiding in organizations. To achieve this aim, the organization should ensure that they have high levels of prosocial motivation, cultural tightness, and uncertainty avoidance. When an organization has low levels of prosocial motivation, low cultural tightness, and low uncertainty avoidance, it has a high potential for knowledge hiding at the highest level in the organization. This happens when uncertainty avoidance and cultural tightness are low; in this situation, individuals have no specific rules regarding appropriate behavior (Taras et al., 2012). Moreover, knowledge hiding happens when norms and sanctions are not provided and when negative behavior is not appropriately sanctioned. In such organizations, individuals are interested in achieving their personal needs, not the organization's needs (Chen et al., 1998; Kagitcibasi, 1997). Individuals, who value competitive advantages, hide their knowledge to protect their individual value (Cerne et al., 2014). For those employees whose knowledge represents a competitive advantage (Milne, 2007; Polanyi, 1997) the knowledge hiding is expected. Moreover, when individuals have low levels of prosocial motivation, their actions do not benefit others. Those employees prioritize only their own needs and willingness (Grant, 2007). Organizational leaders (Aktas et al., 2015) should aim to provide a three-way interaction to prevent knowledge hiding in advance. If an organization will not support prosocial motivation, will not be uncertainty-avoidance-oriented, and will not have rules or sanctions provided by cultural tightness, it can expect high levels of knowledge hiding among its employees. In the light of the above, we propose the following hypothesis: Hypothesis 3. A three-way interaction among prosocial motivation, cultural tightness, and uncertainty avoidance is associated with lower levels of knowledge hiding. The conceptual model with hypotheses is presented in Figure 1. The explanation of the figure is as follows. Regarding that prosocial motivation is a desire to help or contribute to other people (Grant, 2007), we hypothesized that it could negatively influence knowledge hiding. Therefore, we have designed hypothesis 1, stating that prosocial motivation is negatively related to knowledge hiding. The hypothesis 1 is present in Figure 1, marked with H1, with its direct effect on knowledge hiding. To continue with the explanation, cultural tightness is the construct with strong social norms, rules, and sanctions for inappropriate behavior. It is closely linked with uncertainty avoidance, which represents intolerance for ambiguity and uncertainty from formal rules. Therefore, we combined these two national cultural dimensions into a two-way interaction with a hypothesis that low levels of a combination of cultural tightness and uncertainty avoidance lead to the emergence of a high level of knowledge hiding. We marked this two-way interaction with H2. The last and the most important three-way interaction is marked with H3, dealing with a combined influence of all three constructs on knowledge hiding. Figure \:1he conceptual model of the hypothesesregarding the three-wayinteraction s METHODS To test our hypotheses regarding how the three-way interaction of prosocial motivation, cultural tightness, and uncertainty avoidance predicts knowledge hiding, we collected primary data from Chinese and Slovenian employees (n = 376). We obtained data from two samples of working professionals from Slovenia (n = 123) and China (n = 253). We analyzed the data using confirmatory factor analysis (CFA) and with hierarchical regression analyses, including the three-way interaction. Sampling procedure We applied the convenience sampling procedure. A sample contains working professionals from China and Slovenia. For sample 1, in China, data were collected from Chinese employees working in various types of companies. For sample 2, in Slovenia, data were collected from employees working in various sectors and who had various job functions. The country-level score for China cultural tightness is 35.3 and for uncertainty avoidance, it is 30. The country level score for Slovenia cultural tightness is 109.6 and for uncertainty avoidance, is 88. China has a low score on uncertainty avoidance; on the other hand, Slovenia has a high score on uncertainty avoidance (Hofstede, 1983b). Based on the country-level score on cultural tightness, Slovenia has a high score compared to China (Uz, 2015). So far, no research has been made about how social-cultural aspects influence knowledge hiding based on the case of Slovene and Chinese cultures. We approached the survey respondents by sending the web link to employees. We used a web-based online data collection tool and shared the survey links via social media and other online channels (i.e., LinkedIn, Facebook, email address, etc.). Sample 1: China We included Chinese culture and Chinese employees in our research, because of the differences from the compared country (i.e., Slovenia). China is becoming a major player in global business and it is a major change in globalization (Stiglitz, 2007). In this study, only employees with e-mail addresses participated. Participants held a wide variety of jobs. They were not likely to know the purpose of our research because the items of this study were presented in a large-scale questionnaire. Traditional Chinese language was used. For Chinese sample, we collected completed responses from 253 employees who worked in 60 teams. About 62% of the participants were female. The age of participants ranged from 18 to 54 years, and the mean age was 29.89 years (SD = 5.596). On average, participants had more than 4.5 years of work experience (mean work experience = 4.75; SD = 4.57), and had worked for less than 3 years with their current supervisors (i.e., dyad tenure; mean = 2.79; SD = 2.29). Sample 2: Slovenia In Slovenia, we sampled employees who worked in tax and legal departments, government agencies, accounting firms, academia, electrician trades, administration, pharmacies, etc. In the part of research conducted in Slovenia, some employees were contacted via e-mail, some received the questionnaire in paper form, and some received it through social networks (e.g., LinkedIn and Facebook). The questionnaire for Slovenian employees was the same as the one used in China. The respondents were not likely to know the purpose of research, because the items of this study were presented in a large-scale questionnaire. We obtained completed responses from 123 employees in various professions. About 61% of the participants were female. The survey included respondents aged between 18 and 55. The majority (56%) of the employees were between 26 and 35 years old. On average, the participants had more than 6 years of work experience (mean work experience = 6.8; SD = 7.45) and a dyad tenure of 3.8 years (SD = 3.62). Measures For measuring each construct, we used various item scales. We analyzed participants' responses to a level of agreement with behavior rating scale, as follows. The scales for measuring knowledge hiding, uncertainty avoidance and prosocial motivation were assessed on a 7-point Likert-type scale, ranging from 1 (completely disagree) to 7 (I fully agree). The scale for measuring cultural tightness-looseness was assessed on a 6-point Likert-type scale ranging from 1 (completely disagree) to 6 (I fully agree). The participants came from countries with different languages (Slovenian and Chinese), therefore we have used a translation/back-translation procedure (Brislin, 1986). To ensure reliable results and to ensure understanding of questions; we translated the questionnaire from English into their national languages. For the Slovenian sample, we translated the items from English into Slovenian and back into English. We mirrored this approach for the Chinese sample. Knowledge hiding. To measure knowledge hiding, we used the 12-item scale (a = .92) that Connelly et al. (2012) developed. The scale includes items such as "I agreed that I could help my work colleague, but I did not intend to do so"; "I told my coworker that I would help him later, but then I just postponed my help"; and "I told my coworker that I did not know the answer, even though I actually knew." Prosocial motivation. We measured prosocial motivation with the 5-item scale (a = .90) that Grand and Sumanth (2009) developed. The scale includes items such as "I get energized by working on tasks that have the potential to benefit others" and "I like to work on tasks that have the potential to benefit others." Cultural tightness - looseness. We measured cultural tightness-looseness using the 6-item scale (a = .76) developed by Gelfand et al. (2011b). The items measured the cultural perspectives, norms, values, sanctions for inappropriate behavior, and rules in the respondent's country of origin. Sample items include "In my country, there are very clear expectations of how people should act in most situations"; "People agree upon what behaviors are appropriate versus inappropriate in most situations in this country"; and "In my country, if someone acts in an inappropriate way, others will strongly disapprove." Individual perception of uncertainty avoidance. We measured individual perceptions of national cultural dimensions regarding uncertainty avoidance with the four-item scale (a = .98) that Dorfman and Howell (1988) developed. Sample items for uncertainty avoidance include "Leaders expect that employees closely follow the instructions of the procedure"; "Work instructions are important for employees"; and "Standard procedures for employees are very useful in their work." Control variables. We controlled for five control variables, including gender, age, and level of education, average dyad tenure, and country. The reason why we used average dyad tenure as a control variable is that the length of the supervisor-subordinate relationship can influence work perceptions (Fagenson-Eland, Marks. & Amendola, 1997). We also included the country as a control variable to find if country-level cultural differences influenced our results. China was coded with "1", whereas Slovenia was coded with "2". RESULTS The means, standard deviations, and correlations of the pooled Slovenian and Chinese sample, for all studied variables used in the analyses, are shown in Table 1. Reliability indicators - using Cronbach's alphas are also included on the diagonal. Table 1: Means, Standard Deviations, and Correlations of employees within the pooled Slovenian andChinese sample Variable Mean SD 1 2 3 4 5 6 7 8 9 1. Knowledge hiding 1.89 0.99 (.92) 2. Prosocial motivation 5.88 1.02 -.241** (.90) 3. Tightness 4.40 1.05 -.150** .153** (.76) 4. Uncertainty avoidance 5.23 2.47 -.176** .117* .270** (.98) 5. Gender 1.10 1.3 -.108* .022 .028 .826** 6. Age 1.64 1.5 - .088 -.014 -.071 .782** .811** 7. Education 2.93 2.1 - .013 -.083 -.340** .658** .768** .778** 8. Average dyad tenure 3.11 3.10 .101 .009 -.072 -.055 .025 .391** .170** 9. Country 1.40 0.50 .205** -.077 -.379** -.395** -.267** -.250** .061 .146** Note; N = 376 employees from Chinese and Slovenian cultures. Reliability indicators (Cronbach's alphas) are on the diagonal in the parentheses. * p < .05 ** p < .01 The Cronbach's alpha for the constructs of knowledge hiding, prosocial motivation, and uncertainty avoidance is above .90 and shows high reliability. The Cronbach's alpha for cultural tightness is above .76 and is acceptable. From the results in Table 1, we could see high mean values for the constructs of prosocial motivation (5.88), cultural tightness (4.40), and uncertainty avoidance (5.23). The mean results show that values on average are high. On the contrary, from Table 1, we can see that mean value for knowledge hiding is 1.89. Turning to correlations, we found a significant negative correlation between prosocial motivation aid knowledge hiding (-.241; p = .001). Knowledge hiding has also a significant negative correlation with cultural tightness (-.150; p = .001), and with uncertainty avoidance (-.176; p = .001). Prosocial motivation is in a significant positive correlation with cultural tightness (.153; p = .001) and with uncertainty avoidance (.117). Cultural tightness has a significant positive correlation with uncertainty avoidance (.270; p = .001). We applied confirmatory factor analysis using the AMOS 21 software package for ensuring validation of the measurement instruments. We used CFA with the aim of ensuring that the designed model fits the data in a sufficient way. In the analysis of the designed model, we used 26 items to measure the four constructs included in this research (knowledge hiding, prosocial motivation, cultural tightness, and uncertainty avoidance). The results of the CFA—specifying separate factors for each construct—achieved a good model fit (CFI = .910; chi-square = 964.649; RMSEA = .07; df = 266)5. The CFA analysis indicated that all the factor loadings of the four constructs were statistically significant, with a mean standardized loading of .75. This further supports the convergent validity of the constructs. The average standardized loading for knowledge hiding was .70, with a range from .45 to .85. The average standardized loading for prosocial motivation was .79, with a range from .75 to .83. The average standardized loading for cultural tightness was .68, with a range from .45 to .84. The average standardized loading for uncertainty avoidance was .94, with a range from .93 to .95. To test the hypotheses, we used a series of hierarchical regression analyses. In this type of analysis, we test direct relationships and the role of the three-way interaction term (combining three independent constructs—prosocial motivation, cultural tightness, and uncertainty avoidance) in predicting knowledge hiding. The results of the analysis are presented in Table 2. Our analysis is divided into four models. In Model 1, we entered five control variables: gender, age, education, average dyad tenure, and country. We found a significant relationship and influence of average dyad tenure (P = .09; p < .10) and of country (P = .04; p < .05). In Model 2, we entered the prosocial motivation, cultural tightness, and uncertainty avoidance. Through the hierarchical regression analysis, we found a significant negative relationship between prosocial motivation and knowledge hiding (P = -.21; exact p = .000). Therefore, Hypothesis 1 was supported. In Model 3, we entered two-way interactions (prosocial motivation x tightness; prosocial motivation x uncertainty avoidance; and tightness x uncertainty avoidance). We found a significant negative relationship between the two-way interaction of cultural tightness and uncertainty avoidance on knowledge hiding (P = -.23; exact p = .001). Hypothesis 2 was therefore supported. 5 The within-construct residuals were allowed to correlate. Without those modification indices, the model fit would be as follows: CFI .884; chi-square = 1194.537; RMSEA = .08; df = 293. Table 2: Hierarchical regression analyses predicting knowledge hiding Model ] Model 2 Model 3 Model 4 Gender -.05 (.31) -.03 (.55) -.02 (.77) -.00 (.87) Age -.05 (.35) -.05 (.41) -.04 (.42) -.04 (.49) Education .06 (.39) .01 (.85) .00 (.97) .00 (.96) Average dyad tenure .09 1 (.09) 09 1 (08) .09f (.07) 091 ( 08) Country .14 * (.04) .12t (.08) 13t(-06) 13t(.0S) Prosocial motivation .21" (.00) -.23** (.00) .25"(.00) Tightness -.04 (.51) .09 (.18) .04 (.65) Uncertainty avoidance -.OS (.11) -.lilt (.05) -.09f (.08) Prosocial motivation x Tightness IS** ( 00) .16** (.00) Prosocial motivation x Uncertainty avoidance .11 t(-04) .14" (.00) Tightness x Uncertainty avoidance -.23** (.00) -.07 (.43) Prosocial motivation x Tightness x Uncert.Avoid -.21 "(_00) R2 .04 .11 .15 .17 F(df) 3.300 (:■; 3 7:") 5.565 (S; 375) 5.919(11; 375) 6.285 (12; 3 75) .Voies. Team-le*'el n = 3 . "p--.01, *p ; ~p< 10. Standard errors are in parenthes ... H...I tn standardized c «efficients (betas). Values m bold are relevant t( ? the tests of hypotheses. In Model 4, we examined the role of the combined three-way interaction term (prosocial motivation xcultural tightnessxuncertainty avoidance)in stimulatinglmowledgehiding. We analyzed the three-way interaction through the use of tests of simple main-effects, e.g., thaeffectofone varioblg(or set ofuariables(aguoesthe kneko^nother variable (Dawson. &Richter, dad^Ucrk.Ro^mO). Flgu ee 2 :TJte threeibiay icaSerautian nuiong pcosocial mgtivation, tightneas, cod uncerlainiy avoidance peedicting knowledge liding 5,5 -5 - eg a 2,5 - 2 -1- Low PM High PM —o—(1) High Tight, High UA -♦-(2) High Tight, Low UA —□—(3) Low Tight, High UA -■-(4) Low Tight, Low UA Figure 2 shows that the moderation of the combination of low levels of prosocial motivation, low cultural tightness, and low uncertainty avoidance causes employees to engage in more knowledge hiding in the organization. Based on the analysis of simple slopes, we can see that three lines portray a relationship between prosocial motivation, cultural tightness, and uncertainty avoidance with knowledge hiding. The moderation, which supports Hypothesis 3, is significant (p < .01) for both low and high levels of the three-way interaction term. The three-way interaction term of our research was statistically significant. Moreover, the change of F-value between models is in all cases significant (p = .000). The significant results showed us that the combination of low levels of prosocial motivation, low cultural tightness, and low levels of uncertainty avoidance, leads to greater occurrences of knowledge hiding in an organization. In other words, the three-way interaction term moderates and reduces the occurrences of knowledge hiding (the three-way interaction term prosocial motivation x tightness x uncertainty avoidance P = -.21, t = -2.984, p = .003). We controlled for five control variables, including gender (P = -.00, p = .873), age (P = -.04, p = .488), level of education (P = .00, p = .957), dyad tenure (P = .09, p = .081) and country P = .13, p = .059). Only the control variables of dyad tenure and country show influence on the researched relationship. In the following, we describe more in details Figure 3, and Figure 4. Both figures present plotting of two-way interaction, of hypothesis 2. Figure 3 shows the combination of low prosocial motivation and low uncertainty avoidance, which causes employees to be engaged in more knowledge hiding. Figure 4 shows the combination of low prosocial motivation and low cultural tightness, which also causes employees to hide their knowledge from other employees. Based on the analysis of simple slopes, we could see in both figures two lines, which presents a connection with knowledge hiding (e.g. a line of prosocial motivation and uncertainty avoidance; and a line of knowledge hiding on prosocial motivation and cultural tightness). Figure 3 presents a scenario of simple slope analysis referring to the low prosocial motivation and low uncertainty avoidance. Figure 4 presents a scenario of the simple slope analysis referring to the low prosocial motivation and low cultural tightness. Figure 3: The two-way interaction of prosocial motivation with uncertainty avoidance predicting knowledgy hiWing Notes. UA = Uncertainty avoidance, PM = Prosocial motivation. Figu ref:The two-way interaction of prosocial motivation with cultural tightness predicting knowkdge hi&ng Notes. PM = Prosocial motivation, Tight = Cultural tightness. DISCUSSION In the present study, we focused on socio-cultural predictors of knowledge hiding at work. We studied the relationships among social (e.g., prosocial motivation) and cultural dimensions (e.g., cultural tightness and uncertainty avoidance) in predicting knowledge hiding behavior. We collected data from two samples of working professionals in Slovenia and China. As hypothesized, the results showed that prosocial motivation is negatively related to knowledge hiding. Furthermore, as we hypothesized, the results significantly supported the main hypothesis that the three-way interaction among prosocial motivation, perceptions of tight cultures, and uncertainty avoidance would be negatively associated with knowledge hiding at work. Our findings suggest that most knowledge hiding occurs when there are low levels of each element in the three-way interaction term: low prosocial motivation, low cultural tightness, and low uncertainty avoidance. Theoretical contributions The concept of knowledge hiding is still novel and remains largely unexplored (Connelly et al., 2012). In this paper, we make two important theoretical contributions. The first contribution is to the literature of knowledge hiding. We show how socio-cultural aspects influence knowledge-hiding behavior. Research shows that individuals with low levels of all three researched constructs would be more likely to hide knowledge. In previous studies, researchers have focused on knowledge hiding at the individual level and in teams (Babic et al., 2017). Our study extends previous research by focusing on the role of individual perceptions of national culture, specifically examining the interactions among cultural dimensions, its content, and intensity. In particular, low levels of perceived cultural tightness and uncertainty avoidance are related to higher levels of knowledge hiding. Our study thus theoretically contributes to the social-cultural aspects of knowledge hiding. Indeed, with our research, we show how cultural characteristics shape occurrence of knowledge hiding at work. The second theoretical contribution is as a contribution to cross-cultural organizational behavior research (Brewer & Venaik, 2014; Tsui et al., 2007). This study also emphasizes the importance of understanding knowledge hiding in its socio-cultural context of motivations, cultural tightness, and uncertainty avoidance. We find that personal and cultural components play roles in fostering or stifling knowledge hiding. We aimed to conceptualize and empirically validate the proposed relationship (i.e., the three-way interaction among the micro-level perceptions of national culture content - uncertainty avoidance, and strength - tightness, and prosocial motivation) and its association with knowledge hiding. As Cerne et al. (2015), and Babic et al. (2017) have previously shown that increased levels of prosocial motivation are associated with less knowledge hiding, we aimed to empirically test the combined role of social and cultural dimensions. The third contribution is of empirical nature and answers the call for more micro-level cultural research (Taras et al., 2016). With this paper, we strengthen previous research based on a larger sample, different participants and cultures participated. In our research, we have used respondents from two countries and measured their individual perceptions of cultural tightness and uncertainty avoidance. This is a strong push away from cultural stereotyping toward a more genuine understanding of individual behaviors (including knowledge hiding), and the way we perceive the world around us - both in terms of social relations as well as cultural postulates. We would hope to see that our study is contributing to launch the stream of micro-based studies of work behaviors in their social and cultural context (e.g., Brewer & Venaik, 2014; Tsui et al., 2007) as perceived by every single individual. Nations and countries are simply too inaccurate placeholders to understand strength and content of values and the way they shape our behaviors. In fact, stereotyping is a dangerous phenomenon that we urge to avoid. Practical implications Based on results of our study we suggest the following important practical implications for leaders and working professionals in global organizations. They could all benefit from our findings by understanding the joint effect of cultural and social context and the way it shapes knowledge hiding behaviors. Leaders should be aware of detrimental effects knowledge hiding has for organizations, teams, and individuals with reduced quality of relationships, trust, creativity, and innovation (Connelly et al., 2012; Connelly et al., 2014; Cerne et al., 2014; Lewicki, McAllister. & Bies, 1998). As the essence of leadership is to achieve results through others and it is by default a social process, leaders can influence prosocial motivations and behaviors via a variety of approaches, ranging from perspective taking, tools for collaboration, and increasing giving and helping behaviors at work (Bolino & Grant, 2016; Skerlavaj, 2017). An example of an organization that is strengthening prosocial motivations and behaviors of their members is innovation and strategic consultancy IDEO (Amabile, Fisher. & Pillemer, 2014). Second, to shape cultures that would decrease knowledge hiding, organizations need to emphasize individual perceptions of cultural strength (tightness). If an organization has not designed specific rules for appropriate behavior and sanctions for deviant behavior (Aktas, Gelfand. & Hanges, 2015; Chua, Roth. & Lemoine, 2015), damage of knowledge hiding could be even higher for organizations. In order to prevent knowledge hiding, it is important that managers clearly communicate desired organizational behavior and design sanctions for inappropriate behavior (Gelfand et al., 2007; Gelfand, Nishii. & Raver, 2006) during recruitment, selection, onboarding, training and development, as well as rewarding cycles of organizations. Specifically, managers need to pay attention whether knowledge hiding occurs among their employees and should be aware of low levels of the combination of the three-way interaction among the constructs examined in this study. When knowledge hiding happens, the road to recovery is longer and steeper as relations are harmed, creativity decreased, and trust among employees lost (Cerne et al., 2014). Recovery is possible, yet proactive actions are better. Based on research findings, leaders and organizations should set rules of expected organizational behavior and design appropriate sanctions for deviant behavior based on cultural characteristics (Aktas, Gelfand. & Hanges, 2015). Leaders need to signal that knowledge sharing and helping within their organization is a norm (Bock, Zmud, Kim. & Lee, 2005; Cabrera. & Cabrera, 2005; Staples & Webster, 2008). Slightly less equivocal is the element of uncertainty avoidance as it is on one hand associated with higher entrepreneurial activity and on the other hand also with increased knowledge hiding. Hence, as developing collaborative norms and supporting helping behaviors is essential for reducing knowledge hiding, increasing entrepreneurial activity, productivity, higher profitability, and many more beneficial outcomes for both organizations and individuals (according to extensive meta-analysis of Podsakoff, Whiting, Podsakoff. & Blume, 2009), we would suggest practitioners to start there. Limitations and future research directions Despite the aforementioned theoretical and practical contributions, our paper is not without limitations. The first limitation is related to the study's cross-sectional design, which limits our ability to infer causation. Experimental studies should be conducted to achieve a settled, general conclusion. The second limitation of our study is that we cannot make a general conclusion about the researched relationship, because we included only two cultures in our study. Thus, the generalizability of our current findings across countries and cultures is not clear. Consequently, the future direction is to include other cultures in studies, with the aim to draw final conclusions. Researchers can also include other national-culture dimensions along with the prosocial motivation to study these dimensions' effects on knowledge hiding. This is an important opportunity because the construct of knowledge hiding is still novel and hence interesting field of exploration for both researchers and practitioners. An experimental or a longitudinal study could strengthen future research on this topic. Such studies that could tap into either causal influence among the variables, or examine how the variables evolve and influence each other over time, would allow researchers to overcome the potential challenge of omitted variables and to rule out both alternative explanations and potentially recursive relationships among the studied variables. New constructs could also be included in future research. We suggest linking knowledge hiding with artificial intelligence, digitalization, modern technology and the virtual world. It will be interesting to see if the "modern world of digitalization" supports or inhibits knowledge hiding with its new technology. Moreover, a multilevel study of national cultures, organizational cultures, and individual behaviors could represent a promising avenue of taking research on knowledge hiding forward. We suggest that future research studies the national culture and compares it with organizational culture (Davenport & Prusak, 1998; Al-Alawi, Al-Marzooqi. & Mohammed, 2007; Suppiah & Singh Sandhu, 2011). It could be interesting to see how organizational culture with leadership (Oye et al., 2011) influences knowledge hiding. The field of knowledge hiding still has many opportunities to be explored. Organizational culture influences behavior of employees, probably more than general culture. Therefore, in the future, there should be more research made on how organizational culture could influence employees and their behavior. Another limitation is that in our research we did not control for the nationality of respondents specifically (i.e., instead of this, a culture was included as a control variable). However, our limitation opens new research questions for future research. One of the limitations of our study is also heterogeneity of a rather small sample size. We collected data from a sample of employees from two diverse yet specific countries (Slovenia and China). Future research should thus collect data from working professionals, pertaining to a larger number of cultures, preferably in a way that even more cultural bands from the GLOBE study would be covered. We suggest for future research implementing the same research or researching other constructs, but on a larger sample. This would allow further insight into the cultural mechanisms, norms, and values influencing the occurrence of knowledge hiding in organizations (House et al., 2004). We believe that our research provides useful theoretical and practical contributions, and we open a call for more individual (micro level) cross-cultural organizational behavior research. CONCLUSION Our study focused on understanding how individuals make sense of their cultural and social context in deciding whether to hide or not their knowledge at work. This is the first study to combine the prosocial motivation, cultural tightness, and uncertainty avoidance in their joint interaction with knowledge hiding behaviors at work. Based on two independent samples of working professionals from Slovenian and Chinese organizations we found, that the most knowledge hiding happened when individuals experience low levels of prosocial motivation, low cultural tightness, and low uncertainty avoidance. The high levels of knowledge hiding occur when employees are not motivated by the welfare of others, who are inclined to take the risk and who perceive that deviation from cultural norms might not be sanctioned. Leaders and organizations in global environments should be aware of the occurrences of knowledge hiding related to different cultural characteristics when they work abroad or with employees from other cultures. However, they should avoid stereotyping based on countries as cultural placeholders, but rather focus on individual perceptions of their social and cultural context. To prevent knowledge hiding in organizations, leaders should develop organizations, where prosocial motivation and helping behaviors are a strong norm. REFERENCES Aktas, M., Gelfand, M. & Hanges, P. 2015. Cultural tightness-looseness and perceptions of effective leadership. Journal of Cross-Cultural Psychology, 47(2), 1-16. Al-Alawi, A. I., Al-Marzooqi, N. Y. & Mohammed, Y. F. (2007). Organizational culture and knowledge sharing: critical success factors. Journal of knowledge management, 11(2): 22-42. Amabile, T., Fisher, C. M. & Pillemer, J. (2014). IDEO's Culture of Helping. Harvard business review, 92(1-2), 54-61. Ardichvili, A., Page, V. & Wentling, T. (2003). Motivation and barriers to participation in virtual knowledgesharing communities of practice. Journal of knowledge management, 7(1), 64-77. Ardila, R., Gouveia, V. V. & Diógenes de Medeiros, E. 2012. Human values of Colombian people. Evidence for the functionalist theory of values. Revista Latinoamericana de Psicología, 44(3): 105-117. Avolio, B. J. & Bass, B. M. (1995). Individual consideration viewed at multiple levels of analysis: A multi-level framework for examining the diffusion of transformational leadership. The Leadership Quarterly, 6(2), 199-218. Babic, K., Cerne, M., Connelly, C., Dysvik, A., Skerlavaj, M. & Zhang, P. (2017). The Interplay among Prosocial Motivation, Cultural Tightness, and Collectivism on Knowledge Hiding. In Academy of Management Proceedings (Vol. 2017, No. 1, p. 13802). Academy of Management. Batson, C. D. 1987. Prosocial motivation: Is it ever truly altruistic? Advances in experimental social psychology, 20, 65-122. Batson, C. D. 2010. Empathy-induced altruistic motivation. Prosocial motives, emotions, and behavior: The better angels of our nature, 15-34. Berry, J. W. (1995). 20. Psychology of Acculturation. The culture and psychology reader, 457. Berry, J. W. (2002). Cross-cultural psychology: Research and applications. Cambridge University Press. Bock, G. W., Zmud, R. W., Kim, Y. G. & Lee, J. N. (2005). Behavioral intention formation in knowledge sharing: Examining the roles of extrinsic motivators, social-psychological forces, and organizational climate. MIS quarterly, 87-111. Bolino, M. C. & Grant, A. M. 2016. The Bright Side of Being Prosocial at Work, and the Dark Side, Too: A Review and Agenda for Research on Other-Oriented Motives, Behavior, and Impact in Organizations. The Academy of Management Annals, 1-94. Brewer, P. & Venaik, S. 2014. The ecological fallacy in national culture research. Organization Studies, 35(7), 1063-1086. Brislin, R. 1986. The wording and translation of research instruments. In W. Lonner. & J. Berry (Eds.), Field methods in cross-cultural research. Cross-cultural research and methodology series, Vol. 8: 137-164. Thousand Oaks, CA: Sage Publications. Cabrera, E. F. & Cabrera, A. (2005). Fostering knowledge sharing through people management practices. The International Journal of Human Resource Management, 16(5), 720-735. Chan, D. K. S. 1996. Tightness-looseness revisited: Some preliminary analyses in Japan and the United States. International Journal of Psychology, 31(1), 1-12. Chen, C. C., Chen, X.-P. & Meindl, J. R. 1998. How can cooperation be fostered? The cultural effects of individualism-collectivism. Academy of Management Review, 23(2): 285-304. Chua, R. Y. In press. Innovating at Cultural Crossroads How Multicultural Social Networks Promote Idea Flow and Creativity. Journal of Management, doi: 0149206315601183. Chua, R. Y., Roth, Y. & Lemoine, J. F. 2015. The Impact of Culture on Creativity: How cultural tightness and cultural distance affect global innovation crowdsourcing work. Administrative Science Quarterly, 60(2), 189-227. Cialdini, R. B., Kallgren, C. A. & Reno, R. R. 1991. A focus theory of normative conduct: A theoretical refinement and reevaluation of the role of norms in human behavior. Advances in experimental social psychology, 24(20): 1-243. Connelly, C. E. & Kevin Kelloway, E. (2003). Predictors of employees' perceptions of knowledge sharing cultures. Leadership & Organization Development Journal, 24(5), 294-301. Connelly, C. E. & Zweig, D. 2015. How perpetrators and targets construe knowledge hiding in organizations. European Journal of Work and Organizational Psychology, 24(3), 479-489. Connelly, C. E., Ford, D. P., Turel, O., Gallupe, B. & Zweig, D. 2014. 'I'm busy (and competitive)!'Antecedents of knowledge sharing under pressure. KnowledgeManagementResearch&Practice, J2(l): 74-85. Connelly, C. E., Zweig, D., Webster, J. & Trougakos, J. P. 2012. Knowledge hiding in organizations. Journal of Organizational Behavior, 33(1): 64-88. Conner, S. L., Reardon, J., Miller, C., Salciuviene,L.& Aureskaviciene,V. (2017).Culturalantenedentstorhe normative, affective, and cognitive effects of domestic versus foreign purchase behavior. Journal of Business Economics and Management, 18(1), 100-115. Cox, T. H., Lobel, S. A. & McLeod, P. L. (1991). EtTec ts of ethnic group cultural differencur om cooaerativn and competitive behavior on a group task. Academy of management journal, 34(4), 827-847. Cui, Y., Park, H. H. & Paik, Y. (2016). Individual and Organizational Antecedents of Knowledge Hiding Behavior. Journal of the Korean Operations Research andMancgementScinnceSocicty ^^ ^^ •"*!), 29(8), 12151239. Cerne, M., Babic, K., Connelly, C. E. & Skerlavaj, M. 2015. Team-level knowledge hiding, social leader-member exchange, and prosocial motivation. In Academy ofManagementProceedings,2015(1),16302. Cerne, M., Nerstad, C. G. L., Dysvik, A. & Skerlavaj,M. 2014.What goesaroundcomes around: Knowledge hiding, motivational climate, and creativity. Academy of Management Journal, 57(1): 172-192. Cerne, M., Hernaus, T., Dysvik, A., and Skerlavaj, M. (2017). The role of multilevel synergistic interplay among team mastery climate, knowledge hiding, and jobcharacteristicsinstimulating innovativework behavior. Human Resource Management Journal, 27: 281-299. Damanpour, F. & Aravind, D. 2012. Managerial Innovation: Conceptions, Processes, and Antecedents. Management and Organization Review, 8(2): 423-454. Davenport, T. H. & Prusak, L. 1998. Workingknowledge: How organizations managewhattheyknow. Harvard Business Press. Dawson, J. F. & Richter, A. W. (2004). A significance test of slope differences for three-way interactions in moderated multiple regression analysis. RP0422.AstonUniversity, Birmingham,UK. Dawson, J. F. & Richter, A. W. (2006). Probing three-way interactions in moderated multiple regression: development and application of a slope differencetest.Journal of Applied Psychology, 91(4), 917. Dawson, J. F. (2014). Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29(1), 1-19. De Vries, R. E., Bakker-Pieper, A. & Oostenveld, W. (2010). Leadership= communication? The relations of leaders' communication styles with leadership styles,knowledgesharing and leadershipoutcomes. Journal of business and psychology, 25(3), 367-380. Debus, M. E., Probst, T. M., König, C. J. & Kleinmann, M. (2012). Catch me if I fall! Enacted uncertainty avoidance and the social safety net as country-level moderators in the job insecurity-job attitudes link. Journal of Applied Psychology, 97(3), 690. Doney, P. M., Cannon, J. P. & Mullen, M. R. (1998). Understanding the influence of national culture on the development of trust. Academy of management review, 23(3), 601-620. Dorfman, P. W. & Howell, J. P. 1988. Dimensions of national culture and effective leadership patterns: Hofstede revisited. Advances in International Comparative Management, 3(1), 127-150. Earley, C. P. & Mosakowski, E. (2000). Creating hybrid team cultures: An empirical test of transnational team functioning. Academy of Management Journal, 43(1), 26-49. Fagenson-Eland, E. A., Marks, M. A. & Amendola, K. L. 1997. Perceptions of Mentoring Relationships. Journal of Vocational Behavior, 5i(1), 29-42. Fischer, R., Ferreira, M. C., Eveline Maria Leal, A., Redford, P. & Harb, C. 2005. Organizational Behaviour across Cultures: Theoretical and Methodological Issues for Developing Multi-level Frameworks Involving Culture. International Journal of Cross Cultural Management: CCM, 5(1), 27-48. Gelfand, M. J., Erez, M. & Aycan, Z. 2007. Cross-cultural organizational behavior. Annual Review of Psychology, 58, 479-514. Gelfand, M. J., Higgins, M., Nishii, L. H., Raver, J. L., Dominguez, A., Murakami, F., Yamaguchi, S. & Toyama, M. 2002. Culture and egocentric perceptions of fairness in conflict and negotiation. Journal of Applied Psychology, 87(5), 833. Gelfand, M. J., Lim, B. C. & Raver, J. L. (2004). Culture and accountability in organizations: Variations in forms of social control across cultures. Human Resource Management Review, 14(1), 135-160. Gelfand, M. J., Nishii, L. H. & Raver, J. L. 2006. On the nature and importance of cultural tightness-looseness. Journal of Applied Psychology, 91(6), 1225. Grant, A. M. 2007. Relational job design and the motivation to make a prosocial difference. Academy of Management Review, 32(2), 393-417. Grant, A. M. & Sumanth, J. J. 2009. Mission possible? The performance of prosocially motivated employees depends on manager trustworthiness. Journal of Applied Psychology, 94(4), 927-944. Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S. T. E. L. L. A., Nishida, T., Kim, K. & Heyman, S. (1996). The influence of cultural individualism-collectivism, self construals, and individual values on communication styles across cultures. Human communication research, 22(4), 510-543. Hall, D. J. & Saias, M. A. (1980). Strategy follows structure! Strategic Management Journal, 1(2), 149-163. Hofstede, G. 1980a. Culture and organizations. International Studies of Management & Organization, 10(4), 15-41. Hofstede, G. (1980b). Motivation, leadership, and organization: do American theories apply abroad? Organizational dynamics, 9(1), 42-63. Hofstede, G. (1983a). The cultural relativity of organizational practices and theories. Journal of international business studies, 14(2), 75-89. Hofstede, G. (1983b). National cultures in four dimensions: A research-based theory of cultural differences among nations. International Studies of Management & Organization, 13(1-2), 46-74. Hofstede, G. (1993). Cultural constraints in management theories. The Academy of Management Executive, 7(1), 81-94. Hofstede, G. (2001). Culture's consequences (2nd ed.). Beverly Hills, CA: Sage. Hofstede, G., Hofstede, G. J. & Minkov, M. (1991). Cultures and organizations: Software of the mind (Vol. 2). London: McGraw-Hill. Hofstede, G. (2011). Dimensionalizing cultures: The Hofstede model in context. Online readings in psychology and culture, 2(1), 8. House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W. & Gupta, V. (Eds.). 2004. Culture, leadership, and organizations: The GLOBE study of 62 societies. Thousand Oaks, CA: Sage publications. House, R., Javidan, M., Hanges, P. & Dorfman, P. 2002. Understanding cultures and implicit leadership theories across the globe: an introduction to project GLOBE. Journal of world business, 37(1), 3-10. Hui, C. H. & Triandis, H. C. (1986). Individualism-collectivism a study of cross-cultural researchers. Journal of cross-cultural psychology, 17(2), 225-248. Javidan, M., House, R. J., Dorfman, P. W., Hanges, P. J. & De Luque, M. S. (2006). Conceptualizing and measuring cultures and their consequences: a comparative review of GLOBE's and Hofstede's approaches. Journal of international business studies, 37(6), 897-914. Johnson, T., Kulesa, P., Cho, Y. I. & Shavitt, S. (2005). The relation between culture and response styles: Evidence from 19 countries. Journal of Cross-cultural psychology, 36(2), 264-277. Johnson, J. P., Lenartowicz, T. & Apud, S. (2006). Cross-cultural competence in international business: Toward a definition and a model. Journal of International Business Studies, 37(4), 525-543. Kagitcibasi, C. 1997. Individualism and collectivism. Handbook of cross-cultural psychology, 3, 1-49. Kirk, Roger E. (1995). Experimental Design: Procedures for the Behavioral Sciences, Third Edition. Monterey, California: Brooks/Cole Publishing. Lenway, S. A. & Rehbein, K. (1991). Leaders, followers, and free riders: An empirical test of variation in corporate political involvement. Academy of Management Journal, 34(4), 893-905. Leung, K. (1989). Cross-cultural differences: Individual-level vs. culture-level analysis. International Journal of Psychology, 24(6), 703-719. Lewicki, R. J., McAllister, D. J. & Bies, R. J. (1998). Trust and distrust: New relationships and realities. Academy of management Review, 23(3), 438-458. Litvin, S. W., Crotts, J. C. & Hefner, F. L. (2004). Cross-cultural tourist behaviour: a replication and extension involving Hofstede's uncertainty avoidance dimension. International Journal of Tourism Research, 6(1), 29-37. Liu, D., Jiang, K., Shalley, C. E., Keem, S. & Zhou, J. (2016). Motivational mechanisms of employee creativity: A meta-analytic examination and theoretical extension of the creativity literature. Organizational Behavior and Human Decision Processes, 137, 236-263. Luthans, F., Zhu, W. & Avolio, B. J. (2006). The impact of efficacy on work attitudes across cultures. Journal of World Business, 41(2), 121-132. McSweeney, B. (2002). Hofstede's model of national cultural differences and their consequences: A triumph of faith-a failure of analysis. Human relations, 55(1), 89-118. Miller, J. G. (1994). Cultural diversity in the morality of caring: Individually oriented versus duty-based interpersonal moral codes. Cross-Cultural Research, 28(1), 3-39. Milne, P. 2007. Motivation, incentives and organisational culture. Journal of knowledge management, 11(6), 2838. Nonaka, I. 1994. A dynamic theory of organizational knowledge creation. Organization Science, 5(1), 14-37. Oye, N., Salleh, M. & Noorminshah, A. 2011. Knowledge sharing in workplace: Motivators and demotivators. International journal of managing information technology, 3(4), 71-84. Pan, W. & Zhang, Q. 2014. A Study on Motivations of Graduate Students' Knowledge Hiding Based on Wuli-Shili-Renli System Approach. Natural science, 27, 36. Pelto, P. J. 1968. The differences between "tight" and "loose" societies. Society, 5(5), 37-40. Podsakoff, N. P., Whiting, S. W., Podsakoff, P. M. & Blume, B. D. 2009. Individual-and organizational-level consequences of organizational citizenship behaviors: A meta-analysis. Journal of Applied Psychology, 94(1), 122-141. Polanyi, M. 1997. The tacit dimension. Knowledge in organizations, 135-146. Salk, J. E. & Brannen, M. Y. (2000). National culture, networks, and individual influence in a multinational management team. Academy of Management journal, 43(2), 191-202. Schwartz, S. H. 2007. Basic Human Values: theory, methods, and application. Risorsa Uomo. Shane, S. (1995). Uncertainty avoidance and the preference for innovation championing roles. Journal of International Business Studies, 26(1), 47-68. Shane, S. 1993. Cultural influences on national rates of innovation. Journal of Business Venturing, 8(1), 59-73. Shin, D., Hasse, V. & Schotter, A. P. (2015). The Effect of Culture on Expatriate Deployment Levels: Cultural Distance and Tightness-Looseness. In Academy of Management Proceedings, 2015(1), 15471. Staples, D. S. & Webster, J. (2008). Exploring the effects of trust, task interdependence and virtualness on knowledge sharing in teams. Information Systems Journal, 18(6), 617-640. Steers, R. M., Meyer, A. D. & Sanchez-Runde, C. J. (2008). National culture and the adoption of new technologies. Journal of World Business, 43(3), 255-260. Stiglitz, J. E. (2007). Making globalization work. WW Norton & Company. Suppiah, V. & Singh Sandhu, M. 2011. Organisational culture's influence on tacit knowledge-sharing behaviour. Journal of Knowledge Management, 15(3), 462-477. Skerlavaj, M. (2016): Prososial motivasjon: Inspirered av gode gjerninger overfor andre .. og seg selv. In Dysvik, A., Kuvaas, B. and Buch, R. (Eds): Motivasjon, motivasjonsteorier og prestasjonskulturer, Capellen Damm. Skerlavaj, M., Connelly, C. E., Cerne, M. & Dysvik, A. In press. Tell me if you can: The roles of perceived time pressure, prosocial motivation, and perspective taking in predicting knowledge hiding. Journal of Knowledge Management. Taras, V., Kirkman, B. L. & Steel, P. 2010. Examining the impact of Culture's consequences: a three-decade, multilevel, meta-analytic review of Hofstede's cultural value dimensions. Journal of Applied Psychology, 95(5), 888. Taras, V., Steel, P. & Kirkman, B. L. 2012. Improving national cultural indices using a longitudinal meta-analysis of Hofstede's dimensions. Journal of World Business, 47(3), 329-341. Taras, V., Steel, P. & Kirkman, B. L. 2016. Does country equate with culture? Beyond geography in the search for cultural boundaries. Management International Review, 56(4), 455-487. Toh, S. M. & Leonardelli, G. J. 2013. Cultural Constraints on the Emergence of Women Leaders. Organizational Dynamics, 42, 191-197. Triandis, H. C. 1977. Interpersonal behavior. Brooks/Cole Publishing Company Monterey, CA. Triandis, H. C. 1989. The self and social behavior in differing cultural contexts. Psychological Review, 96(3), 506. Triandis, H. C. & Gelfand, M. J. 1998. Converging measurement of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology, 74(1), 118. Triandis, H. C. (2000). Cultural syndromes and subjective well-being. Culture and subjective well-being, 13-36. Triandis, H. C. (2004). The many dimensions of culture. The Academy of Management Executive, 18(1), 88-93. Trompenaars, F. & Hampden-Turner, C. 2011. Riding the waves of culture: Understanding diversity in global business. Nicholas Brealey Publishing. Tsai, W. (2002). Social structure of "coopetition" within a multiunit organization: Coordination, competition, and intraorganizational knowledge sharing. Organization science, 13(2), 179-190. Tsui, A. S., Nifadkar, S. S. & Ou, A. Y. 2007. Cross-national, cross-cultural organizational behavior research: Advances, gaps, and recommendations. Journal of Management, 33(3), 426-478. Uz, I. (2015). The index of cultural tightness and looseness among 68 countries. Journal of Cross-Cultural Psychology, 46(3), 319-335. Waters, J. A. (1980). Managerial skill development. Academy of Management Review, 5(3), 449-453. Wennekers, S., Thurik, R., van Stel, A. & Noorderhaven, N. (2007). Uncertainty avoidance and the rate of business ownership across 21 OECD countries, 1976-2004. Journal of Evolutionary economics, 17(2), 133-160. Williams, J. D., Han, S. L. & Qualls, W. J. (1998). A conceptual model and study of cross-cultural business relationships. Journal of Business Research, 42(2), 135-143. Zhang, X. & Zhou, J. (2014). Empowering leadership, uncertainty avoidance, trust, and employee creativity: Interaction effects and a mediating mechanism. Organizational Behavior and Human Decision Processes, 124(2), 150-164. RESEARCH AND DEVELOPMENT EXPENSES UNDER IFRS MANDATORY IMPLEMENTATION: A VALUE RELEVANCE APPROACH Hasna Chaibi1 Received: June 28, 2017 Accepted: December 15, 2018 ABSTRACT: The paper focuses on the degree to which the accounting treatment of R&D expenditure is stock price informative followingthe adoption oflAS. Therefore, usingrecent data of French listed companies, starting from the year in which IFRS were applied, 2005-2015, the present study examines the value relevance of the different R&D accounting treatments. Unlike evidence regarding the pre-IFRS period in France, we find that the capitalized portion of R&D is not correlated with market values, suggesting that under IFRS mandatory implementation, R&D assets are not value relevant. The expensed portion of R&D is positively related to market values only for manufacturing companies. Accordingly, we conclude that IFRS implementation has implications on the valuation of R&D expenditure by investors in French firms. Keywords: R&D, capitalized R&D, expensed R&D, value relevance, IFRS, France JEL: M41; M48 DOI: 10.15458/85451.67 1. INTRODUCTION Since the mandatory implementation of International Financial Reporting Standards (IFRS) by listed European companies in 2005, the consolidated financial statements are published in accordance with international accounting standards. The main goal is to provide a common accounting language and ensure greater consistency in the presentation of accounting information in response to the growing internationalization of financial markets. In some European countries such as Germany, Austria and Switzerland, the adoption of international standards was voluntary before 2005. This was not the case for French companies, and the possibility of preparing their accounts according to rules other than national standards, Generally Accepted Accounting Principles (GAAP), was not included in the law. It therefore seemed appropriate to study the consequences of the introduction of IFRS on French companies. Our objective is to highlight the effects on market value of accounting treatment for a specific asset class, namely the research and development (R&D) expenses, given it is affected by the mandatory change. Before IFRS adoption in France, the rules impose the 1 Tunis El Manar University, Tunis, e-mail: chaibihasna@gmail.com immediate recognition of R&D expenditure as a cost, unless the R&D project meets certain conditions. The French rules (GAAP) stipulate that R&D expenses can be exceptionally an intangible asset only if they relate to a specific individual project with a real chance of technical success and economic profitability and whose costs can be obviously determinate. Nevertheless, since 1 January 2005, the accounting treatment of R&D expenditure under IAS2 38 becomes different. Actually, the capitalization of R&D expenses, which was an option treatment under French GAAP, has become an obligation under IFRS. The capitalization of R&D expenses is a consequence of the standard's requirements and not manager's choice. As a result, the development phase of an intangible project should be recognized once six criteria are met3. In the accounting literature, the controversy over intangible assets and particularly R&D expenditure has been present in the accounting debate since several decades (Aboody and Lev, 1998; Lev and Sougiannis, 1996; 1999). Numerous studies provide evidence about the relevance of capitalized R&D expenditure during the period before the transition to IFRS. Zhao (2002) shows that R&D expenses are not value relevant, while Oswald (2008) proves little difference between the value relevance of the expensed and capitalized portion of R&D expenditure. However, few studies have investigated the period following the mandatory adoption of IFRS. We can mention the study of Shah et al. (2013) that examines the value relevance of R&D expenditure in the periods before and after IFRS in the UK. More recently, Gong and Wang (2016) estimate the changes in the value relevance of R&D expenses for periods pre-IFRS and post-IFRS adoption in countries that previously mandated immediate expensing against those that allowed optional capitalization of R&D expenditure ones. Indeed, there are studies that have investigated the value relevance of R&D expenses in France. Ding and Stolowy (2003), for instance, reveal the lack of relevance regarding the decision to capitalize R&D expenses in relation to the market value of the French company. Later, Cazavan-Jeny and Jeanjean (2006) provide evidence suggesting that R&D expenditure, in France, are negatively associated with market value. Nevertheless and to the best of our knowledge, there is no study focusing on the post-IFRS period in France. On that basis, we contribute to the R&D accounting literature by adding an empirical study examining the value relevance of R&D assets and expenses after the mandatory transition 2 International Accounting Standards. 3 IAS 38 Development Capitalization Criteria stipulate: "An intangible asset arising from development (or from the development phase of an internal project) shall be recognized if and only if, an entity can demonstrate all of the following: (a) The technical feasibility of completing the intangible asset so that it will be available for use or sale. (b) Its intention to complete the intangible asset and use or sell it. (c) Its ability to use or sell the intangible asset. (d) How the intangible asset will generate probable future economic benefits. Among other things, the entity can demonstrate the existence of a market for the output of the intangible asset or the intangible asset itself or, if it is to be used internally, the usefulness of the intangible asset. (e) The availability of adequate technical, financial and other resources to complete the development and to use or sell the intangible asset. (f) Its ability to measure reliably the expenditure attributable to the intangible asset during its development." to IFRS in France, where companies are required to activate whether the capitalization criteria are met compared to the discretion available in the past. In this regard, Lev and Zarowin (1999) provide that it is necessary to control the industry effects in any study on the R&D costs, because these expenses are specific to the nature of sectors. Companies operating in different sectors have different behaviors in terms of investment in R&D. As a consequence, the second objective of this study is to examine the separate value relevance of R&D across manufacturing and nonmanufacturing sectors in the post-IFRS period. To pursue these objectives, we employ recent data by focusing on the fiscal years from 2005 to 2015 of listed French firms with R&D activity. We find that the capitalized portion of R&D is not related to market values, with no significant differences across manufacturing and nonmanufacturing firms. Nevertheless, the expensed portion of R&D is positively related to market values of manufacturing companies. The remainder of the paper is structured as follows. Section 2 exposes the R&D treatment, related literature and the development of the research hypotheses. Section 3 presents the details of the research methodology, related valuation models and the data selection process. Section 4 provides the results of the empirical analysis and discussion of the main findings. Finally, Section 5 concludes the study. 2. INSTITUTIONAL BACKGROUND, PRIOR LITERATURE, AND HYPOTHESES 2.1. R&D EXPENSES TREATMENT : BEFORE VS. AFTER IFRS ADOPTION The accounting treatment of R&D expenditure is controversial at an international level. For example, International Accounting Standard (IAS 38) permits the capitalization of development expenditures when certain conditions are met, whereas the US GAAP adopts a stricter approach to the issue. Indeed, the Financial Accounting Standards Board (FASB), which initially authorized the activation of R&D expenditure, adopted an approach in October 1974 with SFAS No. 22 (FASB 1974, §12), which requires all R&D expenses are expensed during the period of their commitments. The only exception is SFAS No. 863, which concerns software. The same approach was adopted in 1998 by the Business Accounting Deliberation Council, requiring Japanese firms to cover all their R&D expenditure. However, certain national accounting standards such as the French one, offer some flexibility to allow the capitalization of R&D costs when certain conditions are simultaneously fulfilled, namely: - The product is well defined ; - The feasibility of the product is confirmed ; - The related costs can be identified ; - The costs can be covered through the revenues generated by the exploitation of the project. - The firm intends to market the product. The French regulations concerning the treatment of R&D expenditure have undergone numerous changes for the consolidated financial statements. Since 2005, all listed companies in the European Union (EU) have been obliged to prepare their annual reports in accordance with international standards (IFRS/IAS). The revised IAS 38 distinguishes between a "research phase" and a "development phase". Research costs must be recognized as an expense when incurred. The revised IAS 38 (§55) considers that a company cannot demonstrate the existence of an intangible asset during the search phase of a project that will generate probable future economic benefits. Nevertheless, development costs are recognized as assets, if and only if the company can demonstrate simultaneously a set of conditions. Actually, IAS 38 Development Capitalization Criteria stipulate: "An intangible asset arising from development (or from the development phase of an internal project) shall be recognized if and only if, an entity can demonstrate all of the following: " a. The technical feasibility of completing the intangible asset so that it will be available for use or sale. b. Its intention to complete the intangible asset and use or sell it. c. Its ability to use or sell the intangible asset. d. How the intangible asset will generate probable future economic benefits. Among other things, the entity can demonstrate the existence of a market for the output of the intangible asset or the intangible asset itself or, if it is to be used internally, the usefulness of the intangible asset. e. The availability of adequate technical, financial and other resources to complete the development and to use or sell the intangible asset. f. Its ability to measure reliably the expenditure attributable to the intangible asset during its development." Although the accounting treatment of R&D expenditure under IFRS appears similar to French GAAP, there is an important disparity. Actually, IFRS requires the capitalization of the R&D expenditure which meets specified criteria, while the French GAAP provides an option to capitalize that R&D expenditure. 2.2. RELATED LITERATURE ON THE VALUE RELEVANCE OF R&D EXPENSES AFTER IFRS TRANSITION Even today, the accounting treatment of R&D expenses still a sensitive issue. Most of the empirical studies focusing on the value relevance of R&D costs investigated the period before the implementation of IFRS (e.g. Zhao, 2002; Ding and Stolowy, 2003; Cazavan-Jeny and Jeanjean, 2006). However, research that has examined the issue of the value relevance of R&D expenditure (capitalized or expensed) in considering the post-IFRS effects is limited. At this day, the related major studies can be summarized in Tsoligkas and Tsalavoutas (2011), Shah et al. (2013) and Gong and Wang (2016). Tsoligkas and Tsalavoutas (2011) assess value relevance of R&D in the UK after IFRS mandatory implementation. The results reveal that the capitalized portion of R&D is positively related to market values, which implies that the stock market absorbs these assets to successful projects promising future economic benefits. Nevertheless, they report a negatively correlation between expensed R&D costs and market value under IFRS, supporting the idea that theses portions of assets do not reflect any future economic benefit and should be expensed. Consequently, the transition to IFRS induced implications for the valuation of R&D expenditure in the UK. Afterward, Shah et al. (2013) continue the research of Tsoligkas and Tsalavoutas (2011), limited to the first three years post-IFRS (2006-2008), by extending the sample period to seven years after adopting IFRS, 2005-2011. This study also examines the impact of the size of companies and sectors on the value relevance of R&D during the period between 2001 and 2011 in the UK, by separating the periods before and after IFRS. The results affirm the value relevance of the capitalized R&D costs in the 11 years of the sample period. However, no improvement is recorded at the R&D capitalized value relevance in the post-IFRS period. Large companies have a higher relevance of the R&D expenses capitalized than small firms. However, the sector specification does not appear to have a significant effect on the relevance of R&D costs. Recently, Gong and Wang (2016) test whether the nature of differences between national GAAP and IFRS rises to differential changes in the value relevance of R&D expenses after the adoption of IFRS across nine countries4, covering pre-IFRS and post-IFRS periods during 1997-2012. They find that the value relevance of R&D expenses declines after IFRS adoption in countries that previously mandated immediate expensing or allowed optional capitalization of R&D expenditure. They do not find change in the value relevance of R&D costs for countries that convert from the mandated capitalization of R&D expenditure to IFRS. However, even Gong and Wang (2016) integrate the French context in the group of the countries examined, their study presents a subtle limitation. Actually, they have not available data on capitalized R&D assets. As a result, one cannot draw conclusions about how different accounting treatments of R&D, capitalization versus expensing, affect the value relevance of R&D costs. Therefore, we will try to contribute to the existing literature on accounting by examining whether there is a value relevance of R&D costs (expensed or capitalized) in the post-IFRS period in France for a recent sample of 11 years during 2005-2015. Moreover, this study provides evidence on industry membership impact by investigating potential differences in the relevance of R&D expenditure (expensed or capitalized) between manufacturing companies and nonmanufacturing ones. 2.3. HYPOTHESES DEVELOPMENT Stark (2008) and Wyatt (2008) argue that the adoption of IFRS would decline the value relevance of R&D treatment. The point of view advanced by Stark (2008) is that the 4 The nine countries are Australia, Finland, France, Germany, Netherlands, Norway, Sweden, Switzerland, and the UK. These countries adopted IFRS since 2005. adoption of IFRS would eliminate discretion in the treatment of R&D expenditure, so would remove a useful way that companies communicate information to stock markets. The capitalization of R&D expenditure was treated as a signal to the market to indicate the quality of R&D spending. Under French GAAP rule, opting for capitalization reflects a management decision for the purpose of transmitting a signal on their profitable R&D projects for the market and distinguishes themselves from competitors. Actually, the major difference between pre-IFRS and post-IFRS, in particular IAS 38, is that in the former case, management has discretion to choose to either capitalize or expense the development costs, while in the latter case, managers are required to capitalize R&D expenditure (development costs) once certain criteria are met. In other words, reported R&D expenses are effectively value relevant under the optional capitalization rule, and not in the context of a mandatory adoption. This argument is consistent with that of Wyatt (2008), who suggests that the most discretionary elements of intangible assets are the most relevant, due to discretion in the accounting treatment. As a result, we expect an absence of the value relevance of R&D expenses after the mandatory capitalization rule. H1. R&D expenditure (expensed or capitalized) has no value relevance in the post-IFRS period. Another important point of discussion in accounting literature is whether the impact of value relevance of R&D costs is different across diverse sectors. Examining manufacturing versus nonmanufacturing firms, Shah et al. (2008) find positive and significant effects of R&D expenditure on market value for both sectors. Nevertheless, Shah et al. (2009) reveal clear-cut sector effects, and present that, compared to nonmanufacturing companies, manufacturing ones are likely to adopt an intensive investment in R&D strategy to maximize their market value. Actually, Ho et al. (2005) explain that manufacturing and nonmanufacturing firms are likely to adopt a different mix of R&D and advertising investments because of the differences in their relative impacts. They argue that while nonmanufacturing firms benefit most from advertising investment for value creation purposes, manufacturing ones are likely to choose a more R&D-intensive strategy beside advertising investment to maximize their market value. Later, and by splitting R&D expenditure into capitalized and expensed portions, Shah et al. (2013) report that, in the case of the big companies, capitalized R&D portion is positively and significantly related to market value for both nonmanufacturing and manufacturing sectors. However, when the coefficient for nonmanufacturing sector is significant and negative, it is insignificant for manufacturing sector. On the other hand, capitalized R&D is positively significantly related to market value of the small manufacturing companies, but there is a negative and significant relationship between expensed R&D and market value for nonmanufacturing firms. As a result, one would not draw consistent conclusion, on the bases of the mixed findings above, confirming if manufacturing or nonmanufacturing firms are associated with a higher R&D relevance. H2. The industry membership has no effect on the value relevance of R&D expenditure. 3. RESEARCH DESIGN 3.1. METHODOLOGY To examine the value relevance of accounting figures, we follow most prior studies (Cazavan-Jeny and Jeanjean, 2006; Oswald, 2008; Tsoligkas and Tsalavoutas, 2011; Shah et al., 2013) that have adopted Ohlson valuation model (1995). The idea is based on the principle that firms' accounting numbers are judged to be 'value relevant' if they are significantly related to their market value (Beaver, 2002). Actually, Ohlson considers the market value of a company as function of book value of equity and expected future residual income. Consequently, the fundamental Ohlson (1995) valuation framework will be the bench mark model throughout the study. In its simple form, the market value of a company is represented by a linear function of its book value of equity and net income: MVit = a0 + axEit + a2BV + eit (1) Where, MVit is the market value for company i at time t (the market capitalization of the company), which is measured four months after the end of the year for each company5. Eit is the earnings of company i at time t. BVtstands for book value of equity of company i at time t. £tt is an error term. Barth et al. (2001) highlight that the main advantage of Ohlson model is that earnings and book value of equity can be divided to examine the value relevance of separate accounting numbers. In our study, we follow prior research (Cazavan-Jeny and Jeanjean, 2006; Tsoligkas and Tsalavoutas, 2011; Shah et al., 2013) and we decompose accounting incomes into: (1) earnings before R&D expenses and (2) the amount of expensed R&D. Similarly, we divide book value of equity into (1) the book value before capitalized R&D and (2) the amount of R&D in the assets. Therefore, the impact of capitalized R&D and expensed R&D on market value is isolated from earnings and book value of equity and the following equation is formed: VMit = p0 + piAEit + p2ABVit + & CapRDit + & ExpRDit + Eit (2) Where, MVit is the market value for company i at time t. AEit is the adjusted earnings for firm i in year t, before the processing of R&D expenditure. ABVit represents the adjusted book value of equity for firm i in year t, which is net of capitalized R&D. CapRDit is the annual amount of capitalized R&D costs. ExpRDit is the annual amount of expensed R&D costs. £tt is an error term. In order to control the model for heteroscedasticity problem, MV, AE, ABV, CapRD, ExpRD variables are deflated by the number of outstanding shares. Barth and Kallapur (1996) and Barth and Clinch (2009) show that the number of outstanding shares is an 5 Actually, the information on the valup ofR&D exp vnses and othervalues is only available when the financial ftatementa are puMishedjinamely foug mont hs after tlie cIoc1 ng date of the amounting peeiod. Moreover, the impact is incorporated into market value in the following period. effective proxy to capture the effect of scale6. By this way, we measure all variables of the regressions on a per share basis. For reducing the possibility of industry effects and to test whether there is a significant difference in the value relevance of R&D expenditure across industries (manufacturing firms vs. nonmanufacturing ones), we follow Shah et al. (2009), Tsoligkas and Tsalavoutas (2011) and Shah et al. a2013) by estim ating mod el (2) separately for two sub samples: manufaoturing firms versus nonmanufacturing firms. Similarly, we include a dummy variable in thd mode1 (2) identifying the industry to which the firm belongs, then the model (3) is expressed as follows: MVit = S0 + S± SECit + S2 AEit + S3 AEit X SECit + S4 ABVit + Ss ABV it X SECit + S6 CAP _RDit + + CAP_RDit X SECit + S8 EXP_RDit + Sg EXP _RDit X SECit + eit (3) Where, MVtt is the market valuo for company i at time t, SECit is a dummy variable which is equal to 0 if iirm iis a nonmanufacturing one, and equal to f ifit is t manu.acturing one. AE & is the adjusted earnings for firm i in year t, before she processing cf R&D expenditure. AE^SECit is the adjusted earnings multiplied by the sector dummy variabla. ABVjt represents the adjustod book value of" equity for firm i m year t, which is net of capitalized R&D. ABVitxSECla rep resents the adjusteR book value of" equity mulaiplied by the oector dummy vEriable. CapRDlt is the annual cmount af capitnlczed R&D costs. CapRDit xSECit is the annual amnunt of eapitalized R&D costs multiplied by the sector dummy variable. ExpRDit is the annual amount of expensed R&D costs. ExpRDitxSECit represents the annual amount of expensed R&D costs multiplied by the sector dummy variable. St is an ercor term. All variables fre deflftccl by the number of outstanding shores . The sample containsobservations on multiple times to different companies. As a result, we use panel data, and according to the Hausman test, the fixed effects model seems most appropriate to our data. 3.f. DATA AND SAM PLE SELECTl ON In order to Frseaf the value aoletance m R&Deapenses after ITRSadoption, we use a sampl e he Frepih fir ms listed in tho ^ock Exchang e c^anie over U ^ ar perifd MOi^ Ml^. eha em]:fel"loar siuily focunao on 1:]:s^ ^Srttf Ot!eoffntIfnifSl In atd^n lo l)tloforreL 1fte ano^as, wsenchldf firmo mfinandal. fect:oi" at ftein i^nanafat stcit^vi^^nt^s ane pvbHshoe1 wrthparticuLv aofounsing logutations.Thfn , aan identify fir ms;lfi^.^lt.iR^Ei^iitiv^][i^siWe als o r e m ove somp ani e s dCl at m ^reqeit vpI^s baming: and bo ok vayue of equiey am misse df whtoh give rise to a sample of 36 compantes aaith fiseoc year united after 1 Janucry 2005. 6 Barth and Clinch (2009) try both book value of equity and number of shares as deflator, and they conclude that employing number of outstanding shares as deflator is more effective to examine the scale effects as ompared to book value of equity. Deflating the variables of the models by the number of outstanding shares certainly produces extreme values. In order to mitigate the effect of extreme outliers, we winsorize the sample variables and ratios at the top and bottom one percentile of their respective distributions. To recuperate the necessary data, we directly use the electronic sites of companies to download the annual reports (or references) issued by companies or the website of the World Library of Annual Reports (www.annupedia.com). We collect stock prices to assess the market value of sample firms from the ABC Stock Exchange website. 3.3. DESCRIPTIVE STATISTICS Table 1 presents the sample distribution of the variables included in the study during the period from 2005 to 2015. What really stands out from this table is that expensed R&D costs are, on average, higher than the capitalized R&D expenses. This finding proves that French companies are more likely to expense the costs of R&D than capitalizing them in the period after IFRS mandatory implementation. Moreover, Table 1 reveals that there is a significant disparity in the intensity of R&D expressed by the remarkable difference in standard deviation of the variables CapRD and ExpRD. Table 1: Descriptive statistics of variables over the period 2005-2015 Variable Number of observations Mean Median Std. dev Min Max MV 396 1.975 1.537 3.549 0.002 39.532 AE 396 1.298 0.831 1.905 -0.064 17.206 ABV 396 0.199 0.092 0.405 -0.903 2.249 CapRD 396 0.018 0.000 0.032 0 0.171 ExpRD 396 0.080 0.024 0.111 0.004 0.611 MV is the market value. AE is the adjusted earnings before the processing of R&D expenditure. ABV represents the adjusted book value of equity net of capitalized R&D. CapRD is the annual amount of capitalized R&D costs. ExpRD is the annual amount of expensed R&D costs. All variables are deflated by the number of ordinary outstanding shares. Table 2 exposes the descriptive statistics of both R&D variables by industry from 2005 to 2015. The values of CapRD and ExpRD of manufacturing companies are, on average, significantly higher than in the nonmanufacturing sector, which let us conclude that the manufacturing firms spend more in R&D activity that the nonmanufacturing ones. Table 2: Descriptive statistics of R&D variables by sectors Variables Number of observations Mean Std. dev Manu Non- Manu Non- Manu Non- manu manu manu Difference P>|i| CapRD 297 99 0.045 0.009 0.090 0.015 0.036 0 ExpRD 297 99 0.217 0.110 0.292 0.171 0.107 0.001 CapRD represents the annual amount of capitalized R&D costs; ExpRD represents the annual amount of expensed R&D costs. All variables are deflated by the number of ordinary outstanding shares. Table 3 shows the correlations among the regression variables. All independent variables, adjusted earnings, adjusted book value of equity, capitalized R&D costs and expensed R&D costs, are positively related to the dependent variable, market value of equity. We can also notice that these variables have a value of VIF "Variance Inflation Factor" that is less than 107, which allow us to conclude that our empirical model does not have a potentially problem of multicollinearity. Furthermore, the highest correlation coefficient is 0.6847, and it is between market value of equity and earnings. Table 3: Correlation matrix and VIF Variable MV AE ABV ExpRD CapRD VIF 1/ VIF MV 1.0000 AE 0.6847 1.0000 1.96 0.510 ABV 0.5962 0.4428 1.0000 1.65 0.606 ExpRD 0.2265 0.2594 0.2300 1.0000 1.32 0.757 CapRD 0.1632 0.1318 0.1887 0.1998 1.0000 1.21 0.826 MV is the market value. AE is the adjusted earnings before the processing of R&D expenditure. ABV represents the adjusted book value of equity net of capitalized R&D. CapRD is the annual amount of capitalized R&D costs. ExpRD is the annual amount of expensed R&D costs. All variables are deflated by the number of ordinary outstanding shares. 4. EMPIRICAL RESULTS 4.1. RESULTS FOR THE VALUE RELEVANCE OF R&D AFTER IFRS IMPLEMENTATION We test two different models. Model (1) is employed to estimate the fundamental Ohlson's (1995) model explaining market value of equity by earnings and book value of equity. Table 4 shows that the model is statistically significant under 1% level by using F-test. For this basic model, the coefficient on earnings, E, is statistically close to 2, and the coefficient on capital equity, BV, is roughly equal to 1. Earnings and book value of equity are positively significantly related to market value. This finding is consistent with prior research which 7 A limit suggested by Gujarati (1995) and Kennedy (1998). examined the value relevance of R&D in France prior to 2005 (e.g. Cazavan-Jeny and Jeanjean, 2006; Gong and Wang, 2016), indicating that accounting numbers are strongly associated with share prices (adjusted R2 is relatively high, 65%). Model (2) is employed to examine the value relevance of R&D expenditure for the sample from year 2005 to 2015 (table 4). It is similar to model (1) except that we isolate the effect of both capitalized and expensed R&D on market value of equity. This decomposed model allows us to highlight the portion of capitalized R&D (CapRD ) and expensed R&D (ExpRD). Model (2) has four independent variables, versus two for model (1). The regression is again significant under F-test at 1% level. Overall, adjusted R2 for model (2) is higher than adjusted R2 for basic model (1) (65% versus 71%). This improvement in explanatory power of the regression indicates that reporting of R&D is a significant factor in the statistical explanation of market value of equity. Table 4: Value relevance of R&D over the period 2005-2015 Variable Coefficient t value P>|t| Basic Model 1 E 2.084" 2,9 7 0,004 BV 0.890** 2,61 0,009 Intercept 0.638*** 5.22 0,000 F 144 6.38 0.000 Adj ustedR2 0.65 N 39 6 Decomposed Model 2 AE 2. 912*** 6,0 2 0,000 AB V 0.416*** 3,22 0,000 ExpRD 2.896*** 4,56 0,000 CapRD -1.332 -0,30 0 ,081 Intercept 0.719** 2.5 2 0,002 F 121 7.68 0.000 Adj usted R2 0.71 N 39 (5 MV is the market value for company i at time t. E is the earnings of company i at time t. BV stands for book valueof equity of company i at time t. AEis theadjustedearnings forfirmi in year C before the processing of R&D expendihcre. ABV aepresentstheadjusted book value ofequity fos firm i in year j which is net of capitalized RM). yrae)Ji^jD is the annual amount of te apitahzedR&D coats. ExpRD i sthe annual amount of expenecd D&Dcrsts.All variable. are deflatedby title humbeeof ordinary outrtanding fhares. Themodels are with fixedeffects and undtr control of heteroscedasticity. ModejMVlt = a0 + atElt + a2BV + £lt Modej 2 : MVtt = P0 + P1AEU + P2ABVu + P3 CapRDu + ExpRDu + £lt *** and ar indicate statistical aignificanceatthe 1% and 5% level, respectively. Theestimation r etultsfortheper iod after adoption o ftFRS ( 2005-2015) sh ow that the variables -quitycapital andeeruings .adjust: dftrR &Dca si-ivert(ation to the maiket: ^.i^^s^iii^^^ng that accounting numbers always provide relevant information even after isolating R&D expenditure. Focusing on the main variables of interest (CapRD and ExpRD), table 4 reveals that the coefficient for expensed R&D costs is significant, while it is insignificant for capitalized ones. ExpRD have a positive impact on share price, revealing that, from the investors' point of view, these expenses provide relevant information after the adoption of the international IAS/IFRS standards in France. Unlike previous studies (Shortridge, 2004; Tsoligkas and Tsalavoutas, 2011) where market participants perceive that expensed R&D only contains unsuccessful projects that give no future benefit, but reduces the firm's value in the same way as any other cost, these expenses are well appreciated by investors in French market after IFRS adoption. On the other hand, capitalized R&D costs are not associated with market news. Actually, the results show that CapRD costs are negatively but insignificantly related to MV. Therefore, R&D capitalization has different effects on price compared to any other asset. Actually, one euro increase in any asset is associated with an increase in share price, whereas one euro increase in a R&D asset has not an effect on share price. The insignificant coefficient reported in Table 4 (coef. = -1.332 ; t value = -0.30) means that R&D capitalization is not associated with "good" or "bad" news, suggesting that investors in French companies do not value R&D assets after IFRS implementation, while they did it prior 2005. Actually, Cazavan-Jeny and Jeanjean (2006) report a negative impact of capitalized R&D expenditure on market value for French firms in a period preceding IFRS adopting (1993-2002). Consequently, we can deduce from this non association between CapRD and the market value that, in a period which made mandatory the capitalization of R&D, there is not value relevance of capitalized R&D costs. This result is consistent with our hypothesis H1 implying that R&D reporting does not create value for French firms under IFRS. Our finding is in line with the idea of Stark (2008) arguing that the adoption of IFRS would hinder the value relevance of R&D treatment. Actually, the adoption of IFRS would remove management discretion in the treatment of R&D costs, so eliminating a way that companies use to communicate information to stock markets by choosing between expensing or capitalizing R&D expenditure. Actually since 2005, managers are required to capitalize R&D expenditure (development costs) provided certain criteria are met. In other words, reported capitalized R&D expenses are effectively value relevant under the optional capitalization rule, and not in the context of a mandatory adoption. Our result is also consistent with the idea of Wyatt (2008), suggesting that the most discretionary items of intangible assets are the most relevant due to discretion in the accounting treatment. The finding is in line with Gong and Wang (2016), offering evidence that countries switching from the mandatory expensing or optional capitalization rule to IFRS (the case of France) see decreases in the value relevance of R&D costs as capitalizing R&D expenditure with future economic benefits is mandatory under IFRS. In summary, it is found that R&D expenses disclosure, regulated by the IAS, contains value-relevant information in the expensed portion of R&D and not in the capitalized one. This finding can be explained by the fact that, once the management discretion is constrained by the adoption of the IAS, market participants consider and believe more on expensed R&D costs than on the capitalized R&D expenditure. 4.2. RESULTS FOR THE INDUSTRY MEMBERSHIP EFFECTS ON VALUE RELEVANCE OF R&D Our third model is used to test Hypothesis H2 in order to examine the value relevance of R&D expenses (capitalized or expensed) under the control of industry membership, namely manufacturing and nonmanufacturing. Adjusted R2 are relatively high and very close in both samples (0.66 and 0.67) indicating the same strong relationship between accounting numbers and share prices whatever the industry. In the case of manufacturing firms, the coefficient of expensed R&D costs (ExpRD) is positive and significant, whereas it is insignificant in the case of nonmanufacturing companies. On the other hand, results show that CapRD variable is insignificantly correlated with the company's market value for both nonmanufacturing and manufacturing sectors, and there is no significant difference between their coefficients (Table 5). This indicates that R&D does not generate intangible assets for both industries. Overall, our results partially support Hypothesis H2 stipulating that association between R&D expenditure and market value does not depend on activity area. Actually, significant difference between both sectors only exists in expensed R&D portions, as we observe a value relevance of expensed R&D costs for manufacturing firms against to nonmanufacturing sector. The results are similar to Shah et al. (2013) revealing that R&D expenditure is positively related to market value only for manufacturing sector. Table5: Valur rdevfncetrfR&D - sectoreffects-m ode/3) Manufacturingfirms Nonmanufacturing firms Difference Coefficient tvalue C oeffici ent f value Coefficient ivalue AE 2. 912*** 3.02 1.310*** 4.25 1.602 1.12 ABV 0.416*** 4.22 0.961*** 3.11 -0.*45 -0.96 Exp RD 2.236*** 4.56 3.208 1.13 -0.980* -1.78 CapRD -1.960 -1.14 -1.3 36 -0.98 -0.667 -0.3* Intercept 0.519*°* 5.28 0.931*** 5.33 -0.41* - 6.2t F AdjustedR2 C 8**** 0.66 297 51*** 0.67 99 MV is the market value for company i at time t. AE is the adjusted narnings for firm i in year t, before the processing ofR&D expenditure. Ally represents the adjustedbook value of equityfor firm linyear g whichis net ofcepitalizedR&D.CapKDit theaonuctlamouirt ofeaprtaHzedR&D costs. ExpRDn theanmialamounty0 expensedR&Dcorsa. SECir tdummynarieMe whieh is ec^aDd Oi°firmiisanonmaoufacheringone, andequal tt>s ifit ^r omenuicchiringone. AH vacirblts rrodeflated by fhe numbero^rdinaryoutetandingsliafer.The modelir wifh firedeffectsand under control of heteroscedasticity. Model 3: MVit = S0 + St SECit + S2 AEit + S2 AEa X SECit + S4 ABVit + Ss ABV a X SECit + S6 CAP_RDit + S7 CAP_RDit X SECit + Sg EXP_RDit + S9 EXP_RDit X SECit + % *** and * indicate statistical significance at the 1% and 10% level, respectively. 5. CONCLUSION The subtle difference between French rule (GAAP) and IFRS regarding the accounting treatment of R&D expenditure is that under IAS 38, the development phase of an internal project shall be recognized as asset if six criteria are met. Therefore, the capitalization of R&D expenses, which was an option under French GAAP, has become an obligation under IAS/ IFRS. In this paper, we explore if R&D expenses are value relevant for investors in French companies listed on the SFB 120 after adopting IFRS and for a recent period 2005-2015. The results indicate that, against the concerns that the adoption of IFRS may lead to more value relevant R&D reporting, the mandatorily capitalized portion of R&D expenditure is not value relevant. This was not the case under French GAAP (Cazavan-Jeny and Jeanjean, 2006). Nevertheless, expensed portion of R&D costs is positively associated with the market value of the firms, revealing that investors don't treat the expensed portion of R&D as an association with unsuccessful R&D projects, as it was revealed by almost all previous studies. Following the transition to IFRS, there are sector related valuation differences regarding R&D costs in French companies. Actually, the expensed portion of R&D is significantly value relevant only for manufacturing companies. Relating to our findings, we conclude that our results reject the expectations of Barth et al. (2008) and Ball (2006) that IFRS better reflect companies' fundamentals and support the argument advanced by Stark (2008) and Wyatt (2008) that the adoption of IFRS would hinder the value relevance of R&D reporting. The argument behind is that eliminating the discretion to treat R&D expenditure would remove a useful way by which a company conveys information to the stock market (Stark, 2008). Overall, this research examines the value relevance of R&D expenditure during recent period 2005-20015, which fulfills a gap in the relevant literature for French market. To the best of our knowledge, this is the first study on the value relevance of R&D expenditure involving post-IFRS period in France. A way to research, future studies can develop this issue by examining other interesting markets such as China which in 2007 adopted a set of accounting standards entirely new, Brazil which applied IFRS in 2010, or Canada, India and Korea that have just implemented IFRS in 2011. REFERENCES Aboody, D., Lev, B. (1998). The value relevance of intangibles: The case of software capitalization. Journal of Accounting Research, 36, 161-191. Barth, M. E., Beaver, W. H., Landsman, W. R. (2001). The relevance of the value relevance literature for financial accounting standard setting: Another view. Journal of Accounting and Economics, 31, 77-104. Barth, M. E., Clinch, G. (2009). Scale effects in capital markets-based accounting research. Journal of Business Finance and Accounting, 36, 253-288. Barth, M. E., Kallapur, S. (1996). The effects of cross-sectional scale differences on regression results in empirical accounting research. Contemporary Accounting Research, 13, 527-567. Beaver, W. H. (2002). Perspectives on recent capital market research. The Accounting Review, 77, 453-474. Cazavan-Jeny, A., Jeanjean, T. (2006). The negative impact of R&D capitalization: A value relevance approach. The European Accounting Review, 15, 37-61. Ding, Y., Stolowy., H., (2003). Les facteurs déterminants de la stratégie des groupes français en matière de communication sur les activités de recherche et développement. Finance-Contrôle-Stratégie, 6(1), 39-62. Gong, J.J., Yang, S.I-L (2016). Changes in the value relevance of research and development expenses after IFRS adoption. Advances in Accounting, 35, 49-61. Ho, Y. K., Keh, H. T., & Ong, J. M. (2005). The effects of R&D and advertising on firm value: An examination of manufacturing and nonmanufacturing firms. IEEE Transactions on Engineering Management, 52(1), 3-13. Lev, B., Sougiannis, T. (1996). The capitalization, amortization and value relevance of R&D. Journal of Accounting and Economics, 21, 107-138. Lev B., Zarowin P. (1999). The boundaries of financial reporting and how to extend them. Journal of Accounting Research, 37(2), 353-385. Ohlson, J.A. (1995). Earnings, book values and dividends in equity valuation. Contemporary Accounting Research, 11, 661-687. Oswald, D.R. (2008). The determinants and value relevance of the choice of accounting for R&D expenditures in the UK. Journal of Business Finance and Accounting, 35, 1-24. Shah, S. Z. A., Stark, A. W., Akbar, S. (2008). Firm size, sector and market valuation of R&D expenditures. Applied Financial Economics Letters, 4, 87-91. Shah, S.Z.A., Stark, A. W., Akbar, S. (2009). The value relevance of major media advertising expenditures: Some U.K. evidence. The International Journal of Accounting, 44, 187-206. Shah, S.Z.A., Stark, L., Akbar, S. (2013). International financial reporting standards and the value relevance of R&D expenditures: Pre and post IFRS analysis. International Review of Financial Analysis, 30, 158-169. Stark, A.W. (2008). Intangibles and research-An overview with a specific focus on the UK. Accounting and Business Research, 38, 275-285. Tsoligkas, F., Tsalavoutas, I. (2011). Value relevance of R&D in the UK after IFRS mandatory implementation. Applied Financial Economics, 21, 957-967. Wyatt, A. (2008). What financial and non-financial information on intangibles is value relevant? A review of the evidence. Accounting and Business Research, 38, 217-256. Zhao, R. (2002). Relative value relevance of R&D reporting: An international comparison. Journal of International Financial Management and Accounting, 13(2), 153-174. E/B/R POVZETKI V SLOVENSKEM JEZIKU INTELLECTUAL STRUCTURE OF THE OPEN INNOVATION FIELD: STATE OF THE ART AND A CRITICAL LITERATURE REVIEW KRITIČNI PREGLED LITERATURE IN TRENUTNI OBSEGA ZNANJA NA PODROČJU ODPRTIH INOVACIJSKIH SISTEMOV ALEŠ PUSTOVRH, MARKO JAKLIČ POVZETEK: Prispevek predstavlja pregled koncepta odprtih inovacijskih sistemov in ga umesti v širšo strukturo znanstvenih raziskav s področja inovacij. Koncept odprtih inovacij umesti tudi v zgodovinski okvir razvoja razumevanja inovativnosti ter širšega razumevanje raziskav inovativnosti. Z uporabo različnih bibliografskih metod ocenjuje vpliv koncepta in njegov prispevek k teoriji inovativnosti. Čeprav odprti inovacijski sistemi ne predstavljajo nove paradigme v razumevanju inovativnosti pa je koncept v 15 letih raziskav postal jasno definiran in prinaša pomembne dopolnitve k razumevanju inovacij in pomaga odgovoriti na nekaj ključnih vprašanj, ki so jih postavili raziskovalci inovativnosti. Ključne besede: odprto inoviranje, inovacijski koncepti, inovacijske prakse, inovacijska politika, uporabniške inovacije, bibliometrična analiza FIRM CREDIT RATINGS AND FINANCIAL ANALYST FORECAST PERFORMANCE OCENE KREDITNE SPOSOBNOSTI PODJETIJ IN USPEŠNOST NAPOVEDI FINANČNIH ANALITIKOV NATHAN H. JEPPSON, MATTHEW C. GEISZLER, DAVID F. SALERNO POVZETEK: Avtorji v članku proučujejo razmerje med ocenami kreditne sposobnosti podjetij in uspešnostjo napovedi gibanja dobičkov s strani finančnih analitikov. Avtorji na podlagi postavljene hipoteze ugotovijo, da so visoke ocene kreditne sposobnosti, ki predstavljajo nizko kompleksnost nalog in nizko tveganje, povezano s solventnosjo, povezane z manj razpršenosti in bolj točnimi napovedmi gibanja dobička. Nizke ocene kreditne sposobnosti so povezane z več razpršenosti in manj točnimi napovedmi. Avtorji nadalje ugotavljajo, da kakovost poročil o dobičku deluje kot moderator v proučevanem razmerju. Rezultati študije so uporabni za udeležence na trgu, saj razkrivajo povečano (zmanjšano) vrednost informacij, ki jih vsebujejo ocene finančnih analitikov kadar podjetja dobijo visoke (nizke) ocene kreditne sposobnosti. Ključne besede: ocena kreditne sposobnosti, gibanje dobička, poslovanje podjetja, finančni analitiki WHAT REALLY DEFINES THE PERFORMANCE IN HOTEL INDUSTRY? MANAGERS' PERSPECTIVE USING DELPHI METHOD KAJ RESNIČNO OPREDELJUJE USPEŠNOST V HOTELSKI INDUSTRIJI? MANAGERSKI POGLED Z UPORABO DELPHI METODE VALENTINA BOŽIČ, LJUBICA KNEŽEVIČ CVELBAR POVZETEK: Ta članek združuje empirične študije s področja hotelirstva s poudarkom na razumevanju dejavnikov uspešnosti poslovanja hotelov in jih združuje z mnenji strokovnjakov o najpomembnejših dejavnikih uspešnosti v hotelski industriji. V prvem koraku se članek osredotoča na študijo več kot 60 člankov, razpoložljivih v literaturi s področja hotelirstva in s poudarkom na uspešnosti hotelov. V drugem koraku vključuje metodo Delphi na skupini 10-ih strokovnjakov s področja hotelirstva z namenom preučiti njihovo mnenje o glavnih dejavnikih hotelske uspešnosti. Rezultati so pokazali, da se literatura osredotoča predvsem na preučevanje vpliva praks s področja upravljanja s kadri (HRM), organizacijske kulture, informacijsko-komunikacijskih tehnologij (IKT), blagovne znamke, okoljskih praks in hotelskih zmogljivosti na uspešnost. Vendar pa strokovnjaki niso potrdili, da so ti dejavniki ključni za uspešnost hotelov. Poudarili so pomen lokacije, produktne segmentacije, fleksibilnosti podjetij in sodelovanja med ponudniki turističnih storitev. Ti dejavniki doslej še niso pritegnili pozornosti akademskih krogov in predstavljajo potencialno novo pot za prihodnje raziskave v hotelirstvu in pri razumevanju hotelske uspešnosti. Literatura in strokovnjaki pa skupno ugotavljajo, da so tržna usmerjenost, zadovoljstvo strank, kakovost storitev in poslovni procesi pomembni dejavniki hotelske uspešnosti, ki zahtevajo nadaljnje raziskave in preučevanja. Ključne besede: Delphi metoda, hoteli, uspešnost poslovanja THE INTERPLAY AMONG PROSOCIAL MOTIVATION, CULTURAL TIGHTNESS, AND UNCERTAINTY AVOIDANCE IN PREDICTING KNOWLEDGE HIDING PREUČITEV VPLIVA TROJNE INTERAKCIJE MED PROSOCIALNO MOTIVACIJO, KULTURNO TOGOSTJO IN NAGNJENOSTJO K TVEGANJU NA POJAV SKRIVANJE ZNANJA KATJA BABIČ, MATEJ ČERNE, MIHA ŠKERLAVAJ, PENGCHENG ZHANG POVZETEK: V članku preučujemo družbene in kulturne vidike skrivanja znanja. Raziskava je bila narejena s ciljem poglobiti razumevanje vpliva dimenzij nacionalne kulture in vpliva prosocialne motivacije na pojav skrivanja znanja. Želeli smo ugotoviti ali preučevani konstrukti povečujejo ali preprečujejo skrivanje znanja. S pomočjo kombinacije trojne interakcije med prosocialno motivacijo, kulturno togostjo in nagnjenostjo k tveganju pridemo do glavnih zaključkov, zakaj prihaja do skrivanja znanja v organizacijah. Raziskava je opravljena na primeru slovenske (n = 123) in kitajske kulture (n = 253). Rezultati raziskave pokažejo, da se skrivanje znanja pojavi na visoki ravni, kadar se zaposleni srečajo s kombinacijo nizke ravni trojne interakcije, sestavljene iz: nizke ravni prosocialne motivacije, nizke kulturne togosti in nizke ravni nagnjenosti k tveganju. Povedano drugače, se skrivanje znanja pojavi na visoki ravni (tj. z visoko verjetnostjo), kadar zaposleni niso motivirani s strani ostalih sodelavcev in niso nagnjeni k tveganju ter kadar vedo, da kršenje kulturnih norm ni sankcionirano. Na podlagi dveh opravljenih raziskav, v članku predstavimo tako teoretične kot tudi praktične prispevke za področje skrivanja znanja in medkulturnega organizacijskega vedenja. Ključne besede: prosocialna motivacija, kulturna togost, skrivanje znanja RESEARCH AND DEVELOPMENT EXPENSES UNDER IFRS MANDATORY IMPLEMENTATION: A VALUE RELEVANCE APPROACH IZDATKI ZA RAZISKAVE IN RAZVOJ PRI OBVEZNI VPELJAVI MEDNARODNIH STANDARDOV RAČUNOVODSKEGA POROČANJA: VPLIV NA VREDNOST PODJETJA HASNA CHAIBI POVZETEK: Članek analizira povezavo med računovodsko obravnavo izdatkov za raziskave in razvoj ter gibanjem cen delnic po uveljavitvi Mednarodnih standardov računovodskega poročanja (MSRP). Na podlagi analize podatkov za francoska podjetja, katerih delnice so med letoma 2005 in 2015 kotirale na borzi, članek analizira vlogo različne računovodske obravnave izdatkov za raziskave in razvoj. Rezultati kažejo, da v nasprotju z obdobjem pred uveljavitvijo MSRP, kapitaliziran del izdatkov za raziskave in razvoj ne vpliva na tržno vrednost podjetja. Del, ki je vključen med stroške, je pozitivno povezan s tržno vrednostjo, vendar samo za podjetja iz predelovalnih dejavnosti. Članek tako sklene z ugotovitvijo, da je vpeljava MSRP vplivala na vrednotenje izdatkov za raziskave in razvoj s strani investitorjev v francoskih podjetjih. Ključne besede: raziskovanje, razvoj, izdatki, vrednost podjetja, MSRP, Francija