Risk management and strategy alignment: in ﬂ uence on new product development performance

(1)

Risk management and strategy alignment: in ﬂ uence on new product development performance

Darja Peljhan and Mojca Marc

School of Economics and Business, University of Ljubljana, Ljubljana, Slovenia

ABSTRACT

Drawing on the contingency view of strategic alignment and existing risk management paradigms, a cross-sectional study was conducted to investigate how strategic product-market orientation moderates the impact of risk management system (RMS) on new product development (NPD) performance. A regression model from a sample of 95 companies shows that companies supported by RMS have higher NPD performance than companies without RMS. However, this eﬀect is not stronger for prospectors, i.e. companies most inﬂuenced by NPD.

Contrary to well established belief in the risk management literature and practice, wefind that a state-of-the-art enterprise risk management (ERM) can help prospectors achieve higher NPD performance only up to a point. In fact, prospectors benefit most from relatively less developed (ERM) systems, suggesting that overly extensive control procedures may hinder successful new product development. Based on empirical evidence, the study offers new insights into the interplay between strategy, risk management and NPD performance. From a practical perspective, our findings can serve as a guide to help organisations align their RMS to better support their strategic direction. Such guidance is currently lacking in the academic and professional literature.

ARTICLE HISTORY Received 8 April 2021 Revised 11 November 2021 Accepted 19 November 2021

KEYWORDS

Strategic type; new product development (NPD);

enterprise risk management (ERM); NPD performance

1. Introduction

‘The management of new products is the management of risk’(Long2020). New product development (NPD) managers constantly face the challenge of balancing the risk of failure with the risk of missing good NPD opportunities. Since total risk avoidance is impossible, the conventional wisdom is that high-risk NPD efforts can be largely managed with an appropriate risk management system (RMS; Mu, Peng, and MacLachlan2009). However, there is also concern among practitioners that the growing influence of risk management (RM) is slowing down innovation and entrepreneurship in companies by focusing too much on future risks. Moreover, the academic community has only recently shifted its focus from normative descriptions of RMS components to understanding how RM processes actually work in organisations in different strategic settings.

RMS cover all phases of a comprehensive RM process, from risk identiﬁcation and measurement to risk response. RMS include the management of NPD-related risks, such as the risk of project failure or the risk of turning down good opportunities. Previous studies have shown that organisations which integrate NPD eﬀorts with RM are more likely to improve NPD performance (PwC 2018;

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://

creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.

CONTACT Mojca Marc mojca.marc@ef.uni-lj.si School of Economics and Business, University of Ljubljana, Kardeljeva pl.

17, SI-1000 Ljubljana, Slovenia

https://doi.org/10.1080/09537325.2021.2011192

(2)

Mu, Peng, and MacLachlan2009; Oehmen et al.2014). However, the design (development) of RMS can substantially aﬀect NPD performance management, as it can either increase the ability of organisations to recognise and manage the potential of their innovations, or reduce creativity levels and hinder performance by imposing too many rules and constraining behaviours (Marc, MilošSprčić, and MešinŽagar2018).

With NPD managers facing failure rates as high as 80%, it is important to better understand how RM should function to improve NPD performance. According to Kaplan and Mikes (2016), in risk- taking organisations, the level of tolerated risk (risk appetite) is high and the primary aim of RMS should be to limit overconfidence and established beliefs based on instinct and practice, and move to a more analytical and rational approach to decisions based on risk information. On the other hand, organisations with risk-averse strategies have a low risk appetite and the RMS should not overly discourage them from taking risks. We still do not know much about how RM and strategy mutually affect NPD performance, as the existing literature only offers partial explanations by either addressing RM as part of NPD (Mu, Peng, and MacLachlan2009; Oehmen et al.2014) or focusing on the relationship between strategy type and NPD (Narver, Slater, and MacLachlan2004). Our aim is therefore to investigate the interplay between risk management (RM) and strategy that influences new product development (NPD) performance. No study has yet investigated the boundary con- ditions of this relationship. Based on contingency theory (Otley 2016), we propose that strategy type moderates the impact of RMS development on NPD performance. We empirically test whether companies with an innovation-oriented strategy (such as prospectors), which requires high risk-taking, actually benefit more from RMS development. This is important because a lot resources is dedicated to RMS implementation and understanding better the strategicfit enables a more rational allocation.

We study RMS at the organisational level and examine the stages of development: from the non-existent RMS to the most developed enterprise risk management (ERM) stage. We do not specifically address portfolio risk or the risk of a single NPD project. By estimating a regression model based on (cross-section) survey data from a sample of medium and large companies, we provide empirical evidence with implications for theoretical and practical advances in NPD performance management and RM. Our study contributes to current discussions and empirical analyses in two ways. First, our results highlight potential non-linear effects and drawbacks of RMS on NPD performance in the context of the prospector strategy. Contrary to well established belief in RM literature and practice, wefind that a state-of-the-art ERM can only help prospectors achieve higher NPD performance up to a point. Interestingly, prospectors seem to benefit most from a less sophisticated ERM system design. Second, from a practical perspective, our study shows that understanding thefit between strategy types and RMS designs can be beneficial for managers. For example, more intensive risk measurement seems more appropriate for analysers rather than prospectors. Ourfindings can serve as a guide to help organisations align their RMS to better support their strategic direction. Such guidance is currently lacking in the academic and professional literature.

The rest of the paper is divided into four sections. The following section reviews the literature and develops the central hypothesis. The third and fourth sections present the research method and results. Theﬁnal section discusses the theoretical and managerial implications of our analysis and concludes the paper.

2. Theoretical background

Developing and launching new products or services is a risky endeavour because new products can fail, resulting inﬁnancial losses for a company (Kachouie and Sedighadeli2015). Moreover, there is a risk of missing good opportunities associated with NPD (Ford, Aubert, and Ryckewaert2016). Pre- vious studies have indicated that RM in general contributes to NPD success (Mu, Peng, and MacLa- chlan2009; Oehmen et al.2014) and imply that using RM as a method to improve NPD success rates

(3)

is highly relevant. Managers use RMS as part of management control to coordinate and control the NPD process (Davila,2000; Zirger and Maidique1990), to obtain information to reduce uncertainty in NPD, and to improve cost, schedule, and technical performance, leading to better overall performance of such programms (Oehmen et al.2014). In traditional risk management (TRM), different types of risk are treated separately and assessed within a specific department or function. The distinguish- ing characteristic of more developed RMS, such as ERM, is that different risks are coordinated and treated simultaneously across functions. Given that NPD activities are cross-functional, we argue that a more developed organisation-wide RMS is positively related to organisational NPD performance. However, authors (Amabile 1998; Davila, 2000) also warn that overly sophisticated control mechanisms (such as RM) can hinder the growth of NPD performance by reducing the level of creativity required for product development. The existing literature has not addressed the question of optimal level of RMS development to take full advantage of RM in NPD environment. Our study offersfirst insights into this issue.

Companies implement RMS through different stages, depending on the needs and choices of each company (Paape and Speklé 2012). According to contingency theory, the alignment between an internal support system and its context, which includes strategy type, has significant implications for performance (as in Kober, Ng, and Paul 2007; Lillis and Van Veen-dirks 2008; Otley 2016), including NPD performance (Acur, Kandemir, and Boer2012; Gatignon and Xuereb 1997). Since RMS is part of internal support system, a company’s strategic orientation might present an important contextual boundary condition to the relationship between RMS development and NPD performance. Past research does not provide insights specific to RMS, but like other management systems in an organisation (e.g. Hoque 2004; Kober, Ng, and Paul 2007), the RMS should be aligned with the organisation’s strategic orientation to achieve optimal performance. We currently know that RM indicators must be coherent and linked to a strategy to help managers monitor and evaluate organisational progress towards strategic goals (Arena and Arnaboldi 2014). On the other hand, studies show that RMS is mainly implemented to meet regulatory requirements (Lundqvist2015) and not strategic alignment (Peljhan, MilošSprčić, and Marc 2018), therefore the potential moderation effect of strategy remains ambiguous.

In this paper, we examine strategy based on Miles and Snow’s (1978) typology, which dis- tinguishes between prospectors, defenders, analysers and reactors (see Appendix for a descrip- tion of each type), as this typology remains the most enduring and scrutinised strategy classification system and has been regularly used in evaluating business strategies’impact on organisational performance (Flor and Oltra2013; Lin, Tsai, and Wu2014). Prospectors rely most on NPD to maintain their competitiveness (DeSarbo et al.2005). They are willing to devote the necessary NPD-related efforts and resources to new market opportunities, even if these efforts are risky and may lead to costly failures (Naman and Slevin1993). Prospectors tend to have information needs covering a much broader spectrum than others (Verbeeten2010) and must as well consider many non-quantifiable risks. Prior research shows that such risk-taking organisations require RMS to rationalise decision-making, limit the overconfidence and challenge established beliefs (Kaplan and Mikes 2016). We argue that prospectors benefit more in NPD performance when using RMS compared to other strategy types because they are exposed to a greater number of different risks that they must effectively identify, measure, and manage to achieve their ambitious goals.

To summarise, the general premise we propose in this study is that the impact of the RMS development stage on NPD performance depends on the type of business strategy pursued (Figure 1), and it requires aﬁt between the RMS development stage and the strategy type to achieve higher NPD- related outcomes:

H1. The improvement in NPD performance due to RMS development is greatest for the prospector strategy type.

(4)

3. Research method

We test the proposed hypothesis by estimating regression models using cross-section survey data from a sample of medium and large Slovenian companies (N = 1,117). Identification of hypothesised effects involves comparisons betweenfirms at different stages of RMS development which is only possible if the sample contains enough units at different development stages. Since RMS is not yet widely used in the target population (Berk and Loncarski2011), we applied several procedural steps (Groves et al.2009) to oversample potential RMS users and allow for adequate statistical analysis for the purpose of testing the hypothesised relationships. To account for the sample design (over- sampling of large companies and companies from thefinancial sector), we use sampling weights when reporting the descriptive statistics.

The questionnaire was reviewed by judges, i.e. knowledgeable people in the content area, and tested in a mini-pilot study to check for content validity. We addressed the potential problems due to common-method bias in two ways i) procedurally and ii) statistically (Podsakoff, MacKenzie, and Podsakoff2003). First, we separated proximally and methodologically the measurement of key variables by placing the respective questions far apart in the questionnaire and using different scale formats and anchors. Second, we tested for potential common-method problems by applying the common latent methods factor approach. The test indicates that common-method bias is not likely a serious problem.¹ Furthermore, we took several procedural steps specifically to mitigate the social desirability bias. In the invitation letter to participate and in the introduction to the online survey, we assured anonymity, social neutrality and acceptability of all responses.

Before inviting respondents to participate in the online survey, we contacted likely users of RMS (large corporations, publicly traded companies, and companies in theﬁnancial industry) to gather information about the person responsible for RM. After the reminder email, we called these companies and asked them to respond to the questionnaire. Respondents in large companies were CRO or other individuals indicated as most involved in RM; respondents in medium-sized companies were general managers. We collected 136 responses (12% response rate), but due to missing data on some of the variables, we analysed 95 responses in the regression models. Statistical tests indicate that non-response bias is unlikely to be a problem.

The dependent variablenew product development performance(NPD) is measured with a general measure based on Beard and Dess (1981), so it was expected that the target respondents would be knowledgeable about it. Respondents were asked to rate performance over the past year relative to their competitors on a scale of 1 (well below) to 5 (well above). The independent variablestrategy typeis measured using the scale proposed by Shortell and Zajac (1990)–5% of respondents are classiﬁed as reactors, 15% as defenders, 51% as analysts and 30% as prospectors (see also Appendix–

Figure 1.Research model.

(5)

Excerpt from the questionnaire). We include dummy variables to estimate diﬀerences between strategy types (e.g.DstratPis 1 for prospectors and 0 otherwise).

The variableRMS development(RMD) is measured as an ordinal variable that captures four devel- opmental stages. Previous research oﬀers several approaches to measuring the development of RMS;

all are based on the premise that a complete ERM system is the most advanced stage. Prior research also shows that RM and ERM mean different things to different companies (Arena, Arnaboldi, and Azzone2010), making self-reported development stages problematic. To overcome this, we asked respondents whether ten distinctive and systematic elements of RM (such as policies, procedures, and techniques) were present in their respective companies (using yes/no scales) and then classified the companies into four categories of RMS development according to our own pre- defined rules (see Appendix for details). In addition to the procedural steps to ensure content and construct validity, we find statistical support for construct validity in correlations with other questionnaire items related to RM development. Conceptually, our approach is similar to that of Gordon, Loeb, and Tseng (2009), Grace et al. (2015), and Heimeriks, Schijven, and Gates (2012), who also use yes/no scales. Unlike these studies, however, we do not simply sum the number of RM items present to measure the development of RM. Nevertheless, we ran regressions using a summed score of RMS items to check for results’robustness. The results and conclusions remain essentially the same.

The 10 RM items and the rules for classification are based onfindings from the literature. Lundq- vist (2014) used confirmatory factor analysis to identifyfive core dimensions common to several RM frameworks (e.g. COSO, ISO): (1) an internal environment, (2) risk assessment, (3) risk response, (4) information and communication, and (5) monitoring. For each of thesefive dimensions, we ident- ified the representative elements proposed in COSO (McNally 2013) and ISO 31000 (2018) and obtained a list of 10 items covering allfive dimensions. We then set classification rules based on the premise that a company should use at least one element per dimension to have a ‘system’ (see Appendix for details). Based on these criteria, most sample companies (51%) are classified as stage 2 (TRM), 20% as stage 3 (ERM_less), 18% as level 4 (ERM_more), and 11% as level 1 (no RMS). We treatRMDas a categorical (ordinal) variable and therefore include dummy variables for levels 2–4.

We control for the effects of the following variables: industry of thefirm (indicator variable), size (log number of employees), age (five categories: less than 6 years–3%, 6–10 - 4%, 11–15 - 12%, 16–

Table 1.Descriptive sample statistics for continuous variables.

Variable Obs Mean SD Min Max Weighted Mean

Dependent variable

NPD 121 3.38 0.96 1 5 3.48

Control variables

Size_log 136 5.19 1.49 0.69 8.85 5.01

Unc_tech 124 3.17 1.11 1 5 3.10

Unc_dmnd 124 2.90 0.94 1 5 2.90

Unc_pdct 124 3.09 1.13 1 5 3.14

Unc_cmpt 124 2.86 0.97 1 5 2.88

Unc_supl 124 2.69 0.79 1 4 2.67

Unc_inte 124 3.19 1.03 1 5 3.05

Unc_exte 124 3.14 0.95 2 5 2.99

Own_instit 102 20.51 36.40 0 100 19.52

Own_foreign 101 27.46 41.16 0 100 28.67

Own_mngt 101 16.40 34.10 0 100 22.46

Notes: To account for the sample design, we use the following sampling weights in sample statistics: mediumfinancial companies 0.45, largefinancial companies 0.33, medium non-financial 1.85, and large non-financial 0.80.NPDisnew product development performancemeasured as the performance in the past year relative to competitors on a scale from 1 (well below) to 5 (well above). Size_log is the natural logarithm of the number of employees.Unc_tech,Unc_dmnd, Unc_pdct,Unc_cmpt, Unc_supl,Unc_inte, andUnc_exteare uncertainties on a scale from 1 (low) to 5 (high) regarding technology, demand, products, competitors, suppliers, internal environment, and external environment, respectively.Own_instit,Own_foreign, andOwn_mngt are percentages of institutional, foreign, and managerial ownership, respectively.

(6)

20 - 12%, and more than 20 years– 69%), ownership structure (three variables: the percentages owned by managers, by institutional investors and by foreign investors), seven variables measuring uncertainty of the business environment (based on Gordon and Narayayan1984; Govindarajan1984;

Miller 1997; a 5-point scale where 1 represents ‘low uncertainty’ and 5 represents ‘high uncertainty’), and (geographic) market orientation (four categories: national–34%, regional–25%, Euro- pean–24% and global–17%).Table 1shows the descriptive statistics of the sample for continuous variables.

We test the moderation hypothesis by regressingNPDonstrategy type,RMDvariable, and interaction eﬀects between both variables. For additional insight, we estimate the following regression equation for each strategy type separately:

NPD=a+b₁D stratP+b₂TRM+b₃ERM less+b₄ERM more+ +4

j=2g_j(DstratP×RMDj)+controlvariables+1,

whereNPDis relative performance in new product development;DstratPis 1 for prospectors and 0 otherwise (we replace DstratP with DstratA for analysers, DstratD for defenders, and DstratR for

Table 2.Regression results for prospectors.

Baseline models (no moderation)

Main models (moderation)

Explanatory variables:

(1) Strategy

(2) RMS

(3) Strategy,

RMS

(4) Strategy,

RMS, interaction;

marginal eﬀects

(5) Strategy,

RMS, interaction;

total eﬀects

TRM 0.618 0.651* 0.878*

(0.412) (0.373) (0.463)

ERM less 0.662 0.699* 0.689

(0.457) (0.413) (0.489)

ERM more 1.040** 1.000** 1.285**

(0.464) (0.409) (0.528)

Prospectors 0.765*** 0.755*** 1.030*

(0.224) (0.220) (0.563)

× No RM 1.030*

(0.563)

× TRM −0.536 1.371***

(0.668) (0.514)

× ERM less 0.569 2.288***

(0.760) (0.617)

× ERM more −0.469 1.847***

(0.666) (0.519)

Other strategies

× No RM

× TRM 0.878*

(0.463)

× ERM less 0.689

(0.489)

× ERM more 1.285**

(0.528)

Control variables yes yes yes yes yes

Constant 3.461*** 2.541*** 2.935*** 2.834*** 2.834***

(0.638) (0.761) (0.677) (0.759) (0.759)

Adjusted R² 0.160 0.071 0.190 0.199 0.199

n 95 95 95 95 95

Notes: *, **, and *** denote statistical signiﬁcance at 10%, 5%, and 1% level, respectively; robust standard errors shown in second row.TRM,ERM_less,andERM_moreare dummy variables forRMDstages. Control variables are industry, size, age, three variables measuring ownership structure, seven variables measuring business environment uncertainty, and (geographic) market orientation.

(7)

reactors);TRM,ERM_less, andERM_moreare indicator variables forRMDstages 2, 3, and 4, respectively. A statistically significant positiveγcoefficient confirms the moderation hypothesisH1. The VIF statistics show that multicollinearity does not present a considerable threat to our analyses.

4. Results

Table 2presents the regression results for the prospector strategy type: columns (1)–(3) contain the results of the baseline models (without moderation effects), while columns (4) and (5) contain the estimates of the main model. The results show that prospectors have statistically significantly higher NPD performance than other strategy types, with or without accounting for RMD (.765, p=.001 and .755,p=.001, respectively; columns 1 and 3). In column (2), wefind positive effects of RMD(without controlling for the effect of prospector strategy), indicating thatfirms with a more developed RMS have higherNPDperformance.

Column (3) includes both the effect of prospector strategy and the effect of theRMDstage. Com- pared to (1) and (2), the coefficients in (3) are slightly smaller but statistically significant. This indicates that both variables seem to influence NPD performance independently. The results in columns (4) confirm this. The simple effects (β) of prospectors’strategy andRMDstage are positive (all statistically significant except for less developed ERM).

However, our results do not support the moderation hypothesisH1, as none of the interaction effects (γ) in column (4) is statistically significant. The interaction effect is positive (.569,p=.457) only for less developed ERM, while the effects are negative for both TRM (-.536,p=.425) and more developed ERM (-.469,p=.484). This means that we cannot statistically confirm that theRMDeffect is stronger for prospectors than other strategy types. The total effects of RMD are estimated for prospector and other strategies in column (5) as the sums of marginal effects from column (4). For example, summing the marginal effects for prospectors (1.030), TRM (0.878), and their interaction (−0.536) gives the total

Figure 2.Predicted new product development (NPD) performance for strategic types at diﬀerent stages of risk management (RM) development stage.

(8)

effect for prospectors with TRM (1.371), which indicates that prospectors with TRM have on average a 1.371 higher NPD score compared to other strategies without RM. The total effects are all positive and statistically significant for prospectors and, except one, also for other strategy types.

Analysing the overall effects based on the moderation model (seeFigure 2, upper left panel), we find that prospectors can benefit most from a less developed ERM stage. The predicted level ofNPD is highest, as is the difference from prospectors without RM (1.258,p=.043). The magnitude of the effect corresponds to approximately 1.3 SD ofNPD. Other stages ofRMDare also associated with higherNPD, but to a lesser extent: prospectors with more developed ERM can expect a .817 point higherNPDscore (p=.094), prospectors with TRM only .341 points higher (p=.514)–both compared to prospectors without RMS. Overall, this suggests that prospectors benefit from implementing a more holistic RMS, but perhaps a comprehensive ERM is too limiting. This is consistent with Marc, Miloš Sprčić, and Mešin Žagar (2018), who found a negative effect of ERM on the expected growth rate, indicating that ERM may also constrain afirm’s growth potential by effectively reducing risk.

Figure 2also shows the total effects for other strategy types. In the upper right panel, the graph shows that only at the most advanced ERM level are analysers (statistically significantly) better at NPDthan when they have no RM (1.3 points higherNPDscore,p= .041). At other levels, the differ- ences inNPDfor analysers are positive but not statistically significant. The moderation effect is not statistically significant, meaning that the difference in slopes for analysers and non-analysers is not detected.

The lower left panel shows the results for defenders. The difference between defenders with and without RM is not statistically significant for any of the stages. Moreover, it is positive only for the most advanced ERM stage. The difference is positive for other strategy types at all stages ofRMD but statistically significant only for the highest ERM level (1.14 points higherNPD score,p= .022).

In terms ofNPD, non-defender strategy types seem to benefit more than defenders from the RM development. This difference is highest and statistically significant in the TRM stage: it is negative, meaning that defenders with TRM have a lower NPD score than other strategy types with TRM (0.7 points lower NPD score, p= .052). However, the overall slopes of defenders and others are quite similar, as the negative interaction effects are not statistically significant.

Finally, the bottom right panel shows that reactors in allRMDstages have a statistically significantly lowerNPDscore than others. On the other hand, reactors with ERM have statistically significantly higher NPD scores than those without RM (e.g. in the ERM_more stage, the NPD score difference is 3.22 points,p= .002). The difference from no RM is also positive for other strategy types but is statistically significant only for the ERM_more stage. The increase inNPDperformance with advanced levels of ERM is higher for reactors than for other strategies because the interaction effects are positive (statistically significant only for ERM_more: the difference in increase is 2.30 points inNPDscore,p= .022).

5. Discussion and conclusion

In this paper, we investigated how strategic product-market orientation moderates the impact of RMS development on NPD performance. Speciﬁcally, we expected that companies pursuing a prospector strategy would beneﬁt more than others with a more advanced RMS. However, our data do not provide support for this hypothesis. Nonetheless, our work contributes several empirical insights not previously addressed by theory. Applying the NPD perspective in validating the combined importance of strategy and RM provides a novel, holistic, and much needed cross-disciplinary nexus. Although the interplay between strategy, RM, and NPD is theoretically recognised in each of the three domains, cross-level knowledge is still largely lacking. The contributions and implications of our study can be summarised along the following points.

First, our study indicates that RMS development and strategy type act as independent variables, each individually inﬂuencing organisational NPD performance, but their alignment is limited. The

(9)

literature does not provide empirical evidence on how strategy and RM interact to inﬂuence NPD performance. Although we build theoretical assumptions on the well-established contingency view and existing RM paradigms that call for RMS to be explicitly tailored to support business strategy for superior performance (Otley2016; Davila,2000), ourﬁndings cannot fully support this view.

One explanation could be that managers underestimate the importance of strategy as an input in RM. Moreover, RMS is often created in isolation from strategy and for other motives (e.g. external, such as regulation). Specifically, the common point is probably only the risk identification but not the other elements of RMS (assessment, monitoring, communication, etc.). Perhaps thefirst step in achieving a betterfit between RMS and strategy is integrating KPIs and risk reporting in annual reports. Another explanation could be that organisations focus too much on risk measurement and not enough on management. As confirmed with other management tools (e.g. balanced scor- ecard, budgeting, activity-based costing), pure risk measurement would not be sufficient for risk management because integration is lacking (Cokins 2001). Ticking all the elements of RMS and filling in the tables is not enough. There must be a discussion of content, followed by action, and the appropriate strategic importance must be given to NPD risks. Our research suggests that RM standard-setters should pay more attention to inform RM managers on how they can actually achieve better integration of RMS with strategy.

Second, our research points to possible nonlinear effects and drawbacks of RMS in risk-taking organisations. Specifically, wefind that RMS development can only contribute to prospectors achieving higher NPD performance up to a certain point. One explanation for this could be that an overly detailed, overly structured RMS is a barrier to the creativity and innovation required for successful NPD of prospectors. This explanation is consistent with the traditional view of NPD that successful new products can be hindered by control procedures (Amabile1998; Davila,2000). Indeed, some authors (e.g. Amabile1998) argue that imposing rules and constraining behaviours (e.g. as part of more developed RMS) may hinder the growth of NPD performance by reducing the level of creativity required for product development. Because a highly developed RMS could also limit the freedom and creativity of engineers, academic, professional and consulting organisations should consider how to educate RM managers about the NPD particularities in designing RMS.

Third, this study provides preliminary guidance for practitioners on how to develop RMS based on their strategy. This is a valuable contribution because the existing academic and professional literature does not yet provide managers with sufficient guidelines in this area. Our results suggest that a fully developed ERM may discourage prospectors from accepting riskier NPD projects, which is the core of their strategy. Elements of ERM are still beneficial as this gives a company a more systematic approach to managing NPD risks. However, too many and too detailed RM elements could be over- kill. Our results show that a fully developed ERM is best for analysers, who typically follow prospectors in NPD, meaning that prospectors effectively take most NPD risk. Since a significant portion of analysers’product portfolio is relatively stable, it is easier to maintain with a more developed ERM.

Moreover, thesefirms analytically monitor prospectors’NPD activities and therefore could benefit from the interaction of their NPD and RM approaches. For defenders, NPD is not a critical strategic element, but they can still benefit from using RMS to manage risks that are particularly relevant to their strategy. They operate in a stable environment, but they need RM because they are quite dependent on their narrow scope and changes in the environment can severely impact them. If they already have RM, it is better to have ERM than just TRM. Reactors may or may not have NPD activities; however, they do not have consistent strategic orientation and only respond when forced to by environmental pressures. By bringing more structure, ERM allows them to better manage their response to this stimulus and perhaps partiallyfills the lack of a formal strategic plan- ning process.

There are some limitations to this study. Although we explicitly designed our study to examine Slovenianfirms, caution should be exercised when interpreting our results beyond this scope. The results could be sample-specific and dependent on how we measure the development of RMS. As firms need more time to align RMS with their strategy, our study might be premature in uncovering

(10)

such eﬀects. As in other studies using self-assessed performance (e.g. Acur, Kandemir, and Boer2012;

Davila,2000; Frishammar and Hörte2007; Mu, Peng, and MacLachlan2009), a potential limitation is that performance was constructed as a subjective, self-reported measure that may be affected by perceptual biases (e.g. social desirability). Since non-financial measures (including NPD performance) used and reported by companies are not standardised, objective measurement is difficult and requires questionnaires to be answered by multiple respondents, adding complexity to the study design beyond its expected benefits. Future studies might extend this research by incorporating more objective measures of NPD performance, helping to reduce potential respondent and common-method biases. Finally, our cross-section data represent a snapshot of practice, and we cannot assess the possibility of alternative causal directions between variables. The usual caveat that association is a necessary but not sufficient condition for causality also applies here. Future research might consider the use of longitudinal data to examine the proposed theoretical causal relationships. Specifically, longitudinal case studies could enhance understanding of the underlying relationships found in our study and their dynamics. Notwithstanding the above limitations, the study provides original insight into the interplay between three critical areas forfirm success: new product development, risk management, and strategic orientation.

Notes

1. All untabulated results are available from the corresponding author upon request.

Acknowledgments

Authors would like to thank two anonymous reviewers for the valuable suggestions, and conference participants at the International Management Control Research Conference (Roehampton University, London, UK, 25–26 June 2019) for the comments and guidance.

Disclosure statement

No potential conﬂict of interest was reported by the author(s).

Funding

The authors acknowledge theﬁnancial support from the Slovenian Research Agency (research core funding No. P5- 0364).

Notes on contributors

Mojca Marcis an Associate Professor of Business Economics at the University of Ljubljana, School of Economics and Business (SEB). As a member of the Academic Unit for Management and Organisation, she teaches courses in management accounting and economics of projects, including postgraduate courses at the Faculty of Chemistry and Chemical Engineering. She is a visiting professor at the University of Zagreb, Faculty of Economics and Business and also teaches at University of Rijeka, Faculty of Economics.. She is a member of Management Control Association and Slovenian academy of Management. Her research interests include performance measurement and management systems, risk management systems as part of management control. Her current work includes studying the eﬀects of calculative culture in risk management, social impact measurement and management, and the interplay between sleep and work. She is the Editor-in-Chief of theEconomic and Business Review.

Darja Peljhanis a Full Professor at the School of Economics and Business, University of Ljubljana. She is a member of Academic Unit for Management and Organisation and an associate member of the Academic unit for Accounting and Auditing. She teaches courses on business performance analysis, management control and accounting for managers.

She is a visiting professor at the University of Rijeka, Faculty of Economics and Business and University of Zagreb, Faculty of Economics and Business. Her research interests are in theﬁelds of management control systems and performance measurement and management. She is a member of European Accounting Association, Management Control Associ- ation, American Accounting Association and Slovenian academy of Management. She is a member of the research

(11)

group‘Inﬂuence of governance, organisational learning and knowledge management on contemporary organisations’ funded by the Slovenian Research Agency. She serves as a member of Editorial Team of the Economic and Business Review and Journal of Management and Business Administration Central Europe.

ORCID

Darja Peljhan http://orcid.org/0000-0001-6566-237X Mojca Marc http://orcid.org/0000-0002-6563-0279

References

Acur, N., D. Kandemir, and H. Boer.2012.“Strategic Alignment and New Product Development: Drivers and Performance Eﬀects.”Journal of Product Innovation Management29 (2): 304–318.

Amabile, T. M.1998.“How to Kill Creativity.”Harvard Business Review76: 77–87.

Arena, M., and M. Arnaboldi.2014.“Risk and Performance Management: Are they Easy Partners?”Management Research Review37 (2): 152–166.

Arena, M., M. Arnaboldi, and G. Azzone. 2010. “The Organisational Dynamics of Enterprise Risk Management.” Accounting, Organisations and Society35: 659–675.

Beard, D. W., and G. Dess.1981.“Corporate-Level Strategy, Business-Level Strategy, and Firm Performance.”Academy of Management Journal24: 663–688.

Berk, A., and I. Loncarski. 2011. “Are Exogenous Requirements Suﬃcient to Induce Corporate Risk Management Activities?”Post-Communist Economies23 (1): 119–137.

Cokins, G.2001.Activity-Based Cost Management: An Executive’s Guide. New York: John Wiley & Sons.

Davila, T. 2000. “An Empirical Study on the Drivers of Management Control Systems’ Design in New Product Development.”Accounting, Organisations and Society25: 383–409.

DeSarbo, W. S., C. A. Di Benedetto, M. Song, and I. Sinha.2005.“Revisiting the Miles and Snow Strategic Framework:

Uncovering Interrelationships between Strategic Types, Capabilities, Environmental Uncertainty, and Firm Performance.”Strategic Management Journal26 (1): 47–74.

Flor, M. L., and M. J. Oltra.2013.“An Exploratory Analysis of the Relationship between Absorptive Capacity and Business Strategy.”Technology Analysis & Strategic Management25 (9): 1103–1117.

Ford, S., C. Aubert, and E. Ryckewaert.2016.Reducing the Risk of Failure in New Product Development: Getting it Right at the Front end of Innovation. Cambridge: University of Cambridge, Centre for Technology Management, Institute for Manufacturing.

Frishammar, J., and SÅ Hörte.2007.“The Role of Market Orientation and Entrepreneurial Orientation for New Product Development Performance in Manufacturing Firms.”Technology Analysis & Strategic Management19 (6): 765–788.

Gatignon, H., and J. M. Xuereb.1997.“Strategic Orientation of the Firm and New Product Performance.”Journal of Marketing Research34 (1): 77–90.

Gordon, L. A., M. P. Loeb, and C. Tseng.2009.“Enterprise Risk Management and Firm Performance: A Contingency Perspective.”Journal of Accounting and Public Policy28: 301–327.

Gordon, L. A., and V. K. Narayayan.1984.“Management Accounting Systems, Perceived Environmental Uncertainty and Organisation Structure: An Empirical Investigation.”Accounting, Organisations and Society9 (1): 56–69.

Govindarajan, V.1984.“Appropriateness of Accounting Data in Performance Evaluations: An Empirical Examination of Environmental Uncertainty as an Intervening Variable.”Accounting, Organisations and Society2: 125–135.

Grace, M. F., J. T. Leverty, R. D. Phillips, and P. Shimpi.2015.“The Value of Investing in Enterprise Risk Management.”The Journal of Risk and Insurance82 (2): 289–316.

Groves, R. M., F. J. FowlerJr., M. P. Couper, J. M. Lepkowski, E. Singer, and R. Tourangeau.2009.Survey Methodology. 2nd ed. Hoboken, NJ: Wiley.

Heimeriks, K. H., M. Schijven, and S. Gates.2012.“Manifestations of Higher-Order Routines: The Underlying Mechanisms of Deliberate Learning in the Context of Postacquisition Integration.”The Academy of Management Journal55 (3):

703–726.

Hoque, Z. 2004. “A Contingency Model of the Association between Strategy, Environmental Uncertainty and Performance Measurement: Impact on Firm Performance.”International Business Review13: 485–502.

ISO 31000.2018. Risk Management–Guidelines.https://www.iso.org/obp/ui#iso:std:iso:31000:ed-2:v1:en.

Kachouie, R., and S. Sedighadeli.2015.“New Product Development Success Factors in Prospector Organisations; Mixed Method Approach.”International Journal of Innovation Management19 (4), DOI:10.1142/S1363919615500401.

Kaplan, R. S., and A. Mikes.2016.“Risk Management–the Revealing Hand.”Journal of Applied Corporate Finance28 (1):

8–18.

Kober, R., J. Ng, and B. J. Paul.2007.“The Interrelationship between Management Control Mechanisms and Strategy.” Management Accounting Research18: 425–452.

(12)

Lillis, A. M., and P. M. G. Van Veen-dirks.2008.“Performance Measurement System Design in Joint Strategy Settings.” Journal of Management Accounting Research20: 25–57.

Lin, C., H. Tsai, and J. Wu.2014.“Collaboration Strategy Decision-Making using the Miles and Snow Typology.”Journal of Business Research67: 1979–1990.

Long, D. 2020. Managing Risk in New Product Development.https://gensight.com/managing-risk-in-new-product- development/.

Lundqvist, S. A.2014.“An Exploratory Study of Enterprise Risk Management: Pillars of ERM.”Journal of Accounting, Auditing, & Finance29 (3): 393–429.

Lundqvist, S. A.2015.“Why Firms Implement Risk Governance–Stepping beyond Traditional Risk Management to Enterprise Risk Management.”Journal of Accounting and Public Policy34 (5): 441–466.

Marc, M., D. MilošSprčić, and M. MešinŽagar.2018.“Is Enterprise Risk Management a Value-Added Activity?”E + M Ekonomie a Management (E&M Economics and Management)21 (1): 68–84.

McNally, J. S.2013.The 2013 COSO Framework & SOX Compliance: One Approach to an Eﬀective Transition. Committee of Sponsoring Organisations of the Treadway Commission. https:// www.coso.org/documents/COSO%

20McNallyTransition%20Article-Final%20COSO%20Version%20Proof_5-31-13.pdf.

Miles, R., and C. Snow.1978.Organisational Strategy, Structure and Process. New York: McGraw-Hill.

Miller, K. D.1997. Measurement of Perceived Environmental Uncertainties: Response and Extension. Purdue CIBER Working Papers. Paper 123. Retrieved fromhttp://docs.lib.purdue.edu/ciberwp/123.

Mu, J., G. Peng, and D. L. MacLachlan.2009.“Eﬀect of Risk Management Strategy on NPD Performance.”Technovation 29: 170–180.

Naman, J. L., and D. P. Slevin.1993.“Entrepreneurship and the Concept of Fit: A Model and Empirical Tests.”Strategic Management Journal14 (2): 137–153.

Narver, J. C., S. F. Slater, and D. L. MacLachlan.2004.“Responsive and Proactive Market Orientation and New-Product Success.”Journal of Product Innovation Management21 (5): 334–347.

Oehmen, J., A. Olechowski, C. R. Kenley, and M. Ben-Daya.2014.“Analysis of the Eﬀect of Risk Management Practices on the Performance of New Product Development Programs.”Technovation34: 441–453.

Otley, D. T. 2016. “The Contingency Theory of Management Accounting and Control: 1980-2014.” Management Accounting Research31: 45–62.

Paape, L., and R. F. Speklé.2012.“The Adoption and Design of Enterprise Risk Management Practices: An Empirical Study.”European Accounting Review21: 533–564.

Peljhan, D., D. MilošSprčić, and M. Marc.2018.“Strategy and Organisational Performance: The Role of Risk Management System Development.” In Performance Measurement and Management Control: The Relevance of Performance Measurement and Management Control Research. Studies in Managerial and Financial Accounting, Volume 33), edited by M. J. Epstein, F. H. M. Verbeeten, and S. K. Widener, 65–91. Bingley: Emerald Publishing Limited.

Podsakoﬀ, P. M., S. B. MacKenzie, and N. P. Podsakoﬀ.2003.“Common Method Biases in Behavioral Research: A Critical Review of the Literature and Recommended Remedies.”Journal of Applied Psychology88 (5): 879–903.

PwC.2018.Managing risks and enabling growth in the age of innovation.https:// www.pwc.com/sk/en/risk-in-review- study.html.

Shortell, S. M., and E. J. Zajac.1990. “Perceptual and Archival Measures of Miles and Snow’s Strategic Types: A Comprehensive Assessment of Reliability and Validity.”Academy of Management Journal33: 817–832.

Verbeeten, F. H. M.2010.“The Impact of Business Unit Strategy, Structure and Technical Innovativeness on Change in Management Accounting and Control Systems at the Business Unit Level: An Empirical Analysis.” International Journal of Management27 (1): 123–201.

Zirger, B., and M. Maidique.1990.“A Model of New Product Development: An Empirical Test.”Management Science36 (7): 867–883.

Appendices

Appendix: Excerpts from the questionnaire Strategy typology

Respondents were provided with the following descriptions of four organisational types, and asked to select the organisational type that they thought most closely describes their organisation:

(13)

Note: Respondents scoring 1 or 2 were classified as‘defenders,’respondents scoring 3–5 were classified as‘analysers,’ respondents scoring 6 or 7 were classified as‘prospectors,’and respondents marking X were classified as‘reactors.’

Risk management development (RMD)

Respondents were asked to answer with yes/no the following questions about the characteristics of risk management in their company:

a. Does your company have a written statement of theﬁrm’s risk appetite? (1) b. Is there an oﬃcial risk management policy and procedures in your company? (1)

c. Are there any workshops organised in your company where managers discuss exposures to diﬀerent types of risks and risk management strategies (so-called‘Risk management workshops’)? (2)

d. Is risk managed with an integrated analysis and management of all identiﬁed corporate risks (e.g.ﬁnancial, strategic, operational, compliance, and reporting risks)? (2)

e. Does your company create a Risk Map indicating position of risks the company is exposed to, considering prob- ability of occurrence and signiﬁcance of identiﬁed risk to the business activity? (2)

f. Do you determine correlations and portfolio risks eﬀects of combined risks? (2)

g. Do you determine quantitative impacts risks may have on key performance indicators? (2) h. Do you have a risk response plan for all signiﬁcant events? (3)

i. Is a formal report on risk and risk management submitted to board level at least annually? (4) j. Do you monitor key risk indicators aimed at emerging risks (not past performance)? (5)

Note: the numbers in parentheses refer to the dimensions of risk management (see Section3)–1) internal environment, (2) risk assessment, (3) risk response, (4) information and communication, and (5) monitoring.

Companies are classiﬁed into four stages according to the following rules:

Stage 1: No items.→No RM system

Stage 2: Random items, but not from all dimensions.→TRM; some systematic elements but not a real system Stage 3: One item from each dimension.→ERM; less developed form

Stage 4: More than one item from each dimension.→ERM; more developed form

Performance

Respondents were asked to rate the performance of their company in the past year relative to their competitors on a scale from 1 (well below) to 5 (well above) for new product development. The respondents were also oﬀered the possibility to answer‘not sure.’