New venture team managers (NVT) need to discern the risk perceptions of their VCs as part of their strategy for obtaining future financing. Correct interpretation by the NVT of early signals of increased venture risk perception by the VCs is vital to promote both earlier communication and more open bargaining concerning the nature of the next reinvestment decision while that decision is still fairly open to influence. This study examines the average change in funding per investing firm in the VC syndicate in moving from the first round to the second round contract. We view this measure as one of the few direct indicants of the VCs' change over time in VCs perception of the risk of the venture. We argue that changes in perceived risk of investments in early stage ventures are more than just a function of the performance results of the firm.
Past research into venture capitalism in the United States has focused heavily on issues such as: the characteristics used by VCs in evaluating NVT for possible first round funding; the nature of the first round deal; the relative valuation of VC-backed ventures relative to non-VC ventures; and the nature and value of VC involvement in the VC-backed venture. VC reinvestment decision making represents an intriguing area of research that has received too little attention to date. The reinvestment decision represents an interesting context in which to unobtrusively examine the nature of how VCs perceive the value of their involvement with the ventures they fund. This approach promises to shed new insight into the widely debated issue of the value ofVC involvement in the venture. This decision provides a rich field context for study of the effects of the level of procedural justice in the VCs-NVT relationship. This study exploits these theoretical opportunities by examining changes in average level of investment per VC firm.
VCs self-discipline their investment behavior in risky new ventures through utilizing a sequence of funding rounds. This use of multiple refunding opportunities allows VCs to regularly formalize their updated risk perceptions into equity sharing contracts with the NVT. VCs, especially in the United States, are widely reputed to be highly sophisticated in their risk assessment capabilities. Such competency is especially crucial for early stage (seed and/or first round) investors. We need more integrative theoretical perspectives to better understand the mechanisms by which the risk surrounding a new venture can be assessed. Financial perspectives focused on appropriate performance measures provide an important part of the overall assessment but the softer behavioral sciences have meaningful contributions to make in assisting us to understand the most important elements of the VCs-NVT exchange relationship that may impact risk assessment.
From a financial perspective, we can readily predict that VCs will be very cognizant of the overall performance level of the venture in making their early stage risk change assessments. Whereas more mature firms may be focused more narrowly on annual income measures, understanding the adequacy of performance of new startup firms will require a dual focus on income and sales. Sales levels are of special interest to VCs because their return objectives on investment typically require rapid sales growth over a 3-7 year period. VCs, knowing that several failures will be unavoidable given the overall risk level of their average investment, need to grow at least a few of the firms in their investment pool to a significant sales volume level that provides greater feasibility for large net returns relative to the overall size of their pool. We therefore examine the parallel influence of both sales annual performance levels and income annual performance levels (measured as net income before taxes). Higher levels of each type of performance would be expected to decrease VCs risk perceptions in a similar fashion, ceteris parabis.
From the softer behavioral sciences, learning approaches (Fiol & Lyles, 1985) and procedural justice approaches (Greenberg, 1990; Gilliland, 1993) also provide helpful risk assessment information. One early attempt at bringing learning theory concepts to bear on the VCs-NVT relationship suggests examining different types of interactions separately as these may involve unique learning approaches (Barney, Busenitz, Fiet, & Moesel, 1996). Two categories of VC involvement that have been previously differentiated in the VC-NVT literature are strategic assistance activities and operational assistance activities. One way to conceptualize these from a learning perspective is that strategic assistance processes require more higher order learning on the part of the NVT while operational assistance processes focus more heavily on lower order learning. Higher order learning is long-term in orientation, involves more unique information processing, and is more likely to impact the entire firm. Lower order learning is short-term in orientation, involves more repetitive information processing, and is less likely to involve the entire workforce of the venture. Although these two approaches to NVT learning are relatively distinct we expect parallel effects of both on changes in VCs' risk perceptions. The more the NVT appears capable of absorbing and utilizing each type of assistance offered by VCs, reflecting differing underlying NVT learning capabilities, the more likely it should be that VCs' risk perceptions are decreased, ceteris parabis. VCs actively monitor the NVT learning process after the first round deal because a NVT that initially appeared very capable may prove inadequate in the face of new environmental threats not foreseeable at the time of first round funding. The greater the learning capacity demonstrated by the NVT in the face of new challenges the greater the confidence the VCs will place in the NVT.
Another form of assistance that seems very promising to explore is the degree to which VCs can use conflicts in the VCs-NVT relationship as potential learning tools for the NVT. We label this conflict recognition assistance. By bringing existing or emerging conflicts between the VCs and the NVT out into the open for active discussion and possible resolution, VCs may be understood to be focusing the attention processes of the NVT in the hopes of creating a perceived need for learning within the members of the NVT. High levels of overt conflict increase the salience of this form of VCs-NVT interaction for both the VCs and the NVT. For the NVT, this assistance strategy is designed by the VCs to increase motivation for learning. For the VCs, this type of interaction represents a unique opportunity to see the potential for rapid learning responses in the NVT. Rather than being strictly focused on the venture itself, as with the learning associated with strategic and operational forms of assistance, conflict recognition assistance is focused on learning the needs and interests associated with the VCs-NVT relationship itself. Independent of the ultimate success of the venture, it represents learning the norm formation and change processes required to preserve this relationship. The more responsive and flexible the NVT is in response to the conflict recognition assistance attempts of the VCs, the lower the level of perceived risks which remain unresolved for the VCs. Increases in positive reactions to conflict recognition assistance will tend to lower VCs risk perceptions ceteris parabis.
Procedural justice approaches are becoming increasingly common in the management literature (Greenberg, 1990). Procedural justice refers to the fairness of the means used in a relationship to reach specific ends or outcomes from the relationship. A key assumption of much of this stream of research is that being perceived as acting fairly is likely to lead to long-term positive benefits for the parties in an exchange relationship. One of the key aspects of procedural justice research is the attempt to identify how an interaction can approach balance where at least one of the parties (sometimes both) perceives at least a satisficing threshold level of fairness in the way decisions are made that impact joint outcomes from the relationship. Much of the existing VCs-NVT literature can be productively reinterpreted as a search for ways to articulate principles of mutuality and fair play. Two recent papers have explicitly extended this approach to the VCs-NVT relationship with one focusing on the effects of procedural justice perceptions of investors (Sapienza & Korsgaard, 1994) and one focused on determinants of procedural justice perceptions of managers (Busenitz, Moesel, Fiet, & Barney, 1996). Our approach is similar to the former study in that it examines the effects of procedural justice perceptions rather than determinants but differs in that we examine the NVTs justice perceptions rather than the investors perceptions.
Claiming effects of the procedural justice perceptions of the NVT on the risk perception levels of the VCs may at first appear counterintuitive. Why not directly measure the fairness perceptions of the VCs concerning the VCs-NVT relationship instead? First, it is frequently claimed that the NVT has far less power in the exchange relationship than the VCs which are funding them. Thus the natural justice perception to consider is the view of the more dependent party in the relationship as this party is more likely to experience injustice due to exercise of superior power. Second, this approach assumes that justice perceptions are difficult to mask from the other party in the relationship even where the perceiving party might have some self-interests at stake if the perceptions become known to their exchange partner. In a myriad number of ways the NVT signals their justice perceptions to the VCs, often unintentionally. Third, VCs are deeply interested in the justice perceptions of the NVT because of the impact these have on the overall climate for the relationship. If the NVT perceives actions of VCs to be unjust then all aspects of the VCs-NVT exchange relationship are likely to be adversely affected. Adverse effects such as these will increase the risk perceptions of the VCs relative to the venture.
We take one somewhat conventional approach to procedural justice by following the classification of procedural justice components adapted by Gilliland (1993). Gilliland notes three primary categories in which procedural justice components are classified including formal characteristics, explanation, and interpersonal treatment. We select one representative component from each of the three categories to create a new scale called global justice. This is in contrast to the more typical approach of viewing the components as immediate precursors to a general fairness item. Global justice is designed to reflect the overall breadth of the three correlated procedural justice categories. In more of a departure from extant procedural justice theory, we experiment with a new measure which we label mutuality-based justice. As contrasted to the breadth of the global justice measure, this measure focuses narrowly on a key component of conflict resolution attempts in high conflict situations- the degree to which VCs are perceived by the NVT to stress mutuality of interests in resolving disputes. A focus on disputes or conflict-laden contexts is certainly not new to procedural justice. In the extreme, all procedural justice measures could be interpreted as reactions to some degree of conflict. Also mutuality is an integral part of procedural justice theory, perhaps most appropriately classified under Gillilands interpersonal treatment category. However, the focus on mutuality of interests as a fair remedy for overt conflict situations is sufficiently narrow to suggest special treatment for this measure. That special treatment involves relating this measure to a conflict-centered assistance measure. In sum, higher levels of global and mutuality-based justice perceptions by the NVT will have an important impact on the VCs-NVT relationship. These will inevitably be signaled through to the VCs who will tend to decrease their risk perceptions accordingly, ceteris parabis.
After proposing three different types of factors that can reduce risk perceptions of VCs ceteris parabis, we now examine whether these factors can truly be considered as independent as has been typical of most past research approaches in each of these areas. The noted exception is the efforts of the stream of research focused on valuation of VC involvement. A primary goal of this stream can be understood to be establishing a close link between VC assistance activities and new venture performance levels. This leads naturally to a joint focus on differing involvement or assistance measures and differing performance measures (although often the latter is used as the dependent variable of interest. Sapienza (1992) hinted at the need for a further integration step when he concluded that the value of VC involvement is a function of the intensity and style of the interaction between the lead VC and the NVT. This appears to suggest that assistance levels, justice perceptions (one way to conceptualize style of interaction) and performance (the value of involvement) should be considered together. We formalize this argument and reconceptualize the question to be answered as: how are the risk perceptions of the VCs affected by the joint interaction of these three types of factors?
What if the procedural justice perceptions of the NVT in the VCs-NVT relationship and the reactions of the NVT to the assistance attempts of the VCs were not assumed to be independent in their effects on risk perceptions of VCs (independence is implied by excluding either construct from a regression model)? The popular focus in organizational theory literature on the pervasive effects of culture mirrors on a macro level the potential impact of procedural justice on a dyadic level. The degree of effort put forth by the VCs to maintain a perception of fairness in their relationship with the NVT is reflected in the level of trust reflected back at the VCs from the NVT, the commitment of the NVT to making the relationship a quality one, and the enthusiasm of the NVT for continuing to involve the VCs in future rounds of funding. Most critically, the justice perceptions will moderate the receptivity of the NVT to receiving various forms of assistance from the VCs and their eagerness to learn from such assistance. The shortsightedness of VCs who fail to establish positive justice perceptions in their NVTs by exploiting their power advantages in the exchange relationship is returned to their disadvantage later. When the NVT requires assistance the VCs may be capable of providing the assistance but learning will only occur if the NVT is also receptive to the information embedded in the assistance. In other words, the NVT can become incapable of perceiving the value of the message because of past experience with the messenger. High positive reactions to various types of assistance from the VCs are therefore unlikely unless the level of procedural justice perceived by the NVT is also high.
How does the level of venture performance enter into the joint relationship between justice perceptions and reactions to VC assistance? Assuming the validity of our previous arguments that procedural justice perceptions establish the necessary conditions for effective learning to result from assistance of the NVT by VCs, will VCs always utilize high assistance strategies with their NVT whenever reactions from the NVT indicate that the efforts of the VCs to be procedurally just have been effective? Here economic theory appears to provide a straightforward contingency approach. Where assistance by VCs is capable of producing above normal value (the necessary condition of high procedural justice), the riskiness of future assistance efforts with the venture will be largely predictable from the performance covariation with past assistance efforts. If above normal assistance efforts have led to above normal performance levels in the past then this pattern will be expected in the next funding round as well. If above average assistance intensity efforts have been associated with below normal performance results (even in the supportive climate associated with high procedural justice) then ventures with above average assistance expectations will be viewed as excessively costly. Such ventures will therefore be viewed as relatively risky compared to lower assistance intensity investments and the average reinvestment level per VC in the venture should be reduced. We therefore propose a series of positive 3-way interactions between these three groups of factors in predicting reduced risk perceptions of VCs.
Based on this commonly shared interaction logic, we propose six hypotheses. Each of these is based on the interaction of matching variants of these three types of factors. Three hypotheses involve the sales performance variable (H1, H2, and H5) while another three involve parallel hypotheses for the income performance variable (H2, H4, and H6). The three types of hypotheses involve two global assistance measures (strategic assistance and operational assistance) each matched with a global measure of justice (global justice). The other type of hypothesis matches an assistance measure (conflict recognition assistance) and a justice measure (mutuality-based justice) based on the narrow range of contexts in which overt or apparent conflict clearly exists between the interests of the VCs and the interests of the NVT. The use of parallel hypotheses helps to better determine the extent to which significant results or particular forms of the relationship are at least as likely to be spurious as systematic.
H1: Positive NVT reactions to the strategic assistance of VCs will be associated with decreases in VCs' risk perceptions where both VCs are perceived by the NVT to be acting in a globally just manner in the VCs-NVT relationship and venture average sales performance levels are above average for the population of VC-funded first round ventures.
H2: Positive NVT reactions to the strategic assistance of VCs will be associated with decreases in VCs' risk perceptions where both VCs are perceived by the NVT to be acting in a globally just manner in the VCs-NVT relationship and venture average income performance levels are above average for the population of VC-funded first round ventures.
H3: Positive NVT reactions to the operational assistance of VCs will be associated with decreases in VCs' risk perceptions where both VCs are perceived by the NVT to be acting in a globally just manner in the VCs-NVT relationship and venture average sales performance levels are above average for the population of VC-funded first round ventures.
H4: Positive NVT reactions to the operational assistance of VCs will be associated with decreases in VCs' risk perceptions where both VCs are perceived by the NVT to be acting in a globally just manner in the VCs-NVT relationship and venture average income performance levels are above average for the population of VC-funded first round ventures.
H5: Positive NVT reactions to the conflict recognition assistance of VCs will be associated with decreases in VCs' risk perceptions where both VCs are perceived by the NVT to be acting in a mutually just manner in the VCs-NVT relationship and venture average sales performance levels are above average for the population of VC-funded first round ventures.
H6: Positive NVT reactions to the conflict recognition assistance of VCs will be associated with decreases in VCs' risk perceptions where both VCs are perceived by the NVT to be acting in a mutually just manner in the VCs-NVT relationship and venture average income performance levels are above average for the population of VC-funded first round ventures.
Top of Page
Previous Page Main Page Next Page
Return to 1996 Topical Index
© 1997 Babson College All Rights Reserved
Last Updated 4/5/97 by Cheryl Ann Lopez
To sign-up for the Center for Entrepreneurial
Studies' publication lists,
please register with the Entrepreneurship WebTeam.