Designing Incentives for startup teams: Form and timing of equity contracting.
Entrepreneurial teams assign equity positions in their startups using a term sheet that details equity splits and conditions (such as delayed or contingent vesting) for being granted those splits. It is conventional wisdom in the entrepreneurial press that equal splits are poor choices (Wasserman 2012, Moyer 2015). The conventional logic is that by not connecting rewards to contribution level equal split contracts can encourage free-riding behaviors. We experimentally test this conventional wisdom, among other entrepreneurial contracting hypotheses. Our results confirm the relationship between equal splits contracting and depressed effort and contribution, but suggest a different causal sequence relative to conventional wisdom. Rather than the contract form being the primitive and the behavior the derived consequence, our results suggest the reverse.The first order differences in contract performance are driven by the sorting of high contributors into non-equal contracts and of low contributors into equal contracts and not by the differences in incentives.
Specifically,our data suggest three behavioral types —high contributors, conditional contributors and low contributors—and these behave differently in negotiations and perform differently even under identical contracting regimes and when faced with different partner behaviors.
In our data, equal contracts attract low contributors. Relative to the other groups,low contributors exhibitloweffortin vesting contracts (where the penalty for free-riding is mild) and even inproportional contracts (where each partner is held fully accountable for his/her effort investment).The second group in our data, conditional contributors exhibit a preference for vesting contracts.Conditional contributors exhibitnear-maximum effort with contribution-proportional splits, lower effortslightly in vesting contracts and lower effortsubstantially in equal contracts, relative to the fully proportional case.They also increase or reduce effort depending on partner behavior. The third group in our data, high contributorsprefer proportion-based contracting to other contract forms and investnear-maximum effort regardless of the contract. Given this near-constant behavior, their contract preferences maximize their individual gain.
While high contributors are not affected by the contract form, low and conditional contributors do exhibit within-type effort changes. These within-type changes are difficult to unravelunambiguouslybut there is evidence that they, again, are driven by factors other than incentive strength. If incentive strength is the dominant factorwe should observe strong performance differences between contract forms even when these are imposed exogenously rather than being selected endogenously by the team. We test this hypothesis in an additional experimental treatment andsee the opposite, that 75% of the effort gap between contracts disappears when contracts are imposed externally.
An implication of these results in practice isthat personality type is the dominant, and incentive strength a secondary, consideration in both effort expended and value generation. Further, for a given mix of personality types, however coalesced, the contract form can make a difference. While a low-powered incentive contract may be sufficient if everyone on the team is a high contributor, a high-powered contractcan improve contributions if conditional or low contributors are present.
Can performance of low-powered incentive contracts be improved?If the key reason for the low performance of equal contracts is the inability to signal a high effort norm, and this drives away high contributors, we may be able to improve performance of equal contracts by changing the signaling process.
In particular, expressed preferences for contract types may be one, but not the only way to signal future effort level. If in addition to signaling their type by the contract forms, teams could exchange signals attesting to their commitment to the joint project, equal contracts may be able to generate higher effort and greater value, relative to the upfront contracting. To investigate this possibility we examine a scenario in which contracts are selected after an initial round of contributions rather than being chosen upfront. This also has an expression in real entrepreneurial teams; frequently the equity terms are not negotiated until part way downstream in the innovation process.
As hypothesized, in this scenariocontracts’ performances move closer together, in fact the effort gap between equal and non-equal contracts is reduced by two thirds. This result is driven mainly by a change in negotiation strategies of teams. In particular, equal contracts are only accepted when initial effort levels are high. That is, initial effort is asalient signal, used by teams as a screening device to determinewhether or not weak or strong incentives are needed, and this screening process prevents low contributors from sorting into low-powered incentive contracts.
In sum, our research suggests that founding team members have stable personality types. With upfront contracting, the contract offers are the only signal of a type and there arefew barriers for free-riders to select into equal division contracts. As a result, high contributors learn to avoid equal division, leading to poor performance of equal contracts. But, with delayed contracting the initial effort levels are another signal team members can send about who they are, and this reduces the salience of the contract preference signal, improving the effort and value generated in equal contracts.These results suggest that entrepreneurial teams should pay more attention to personality type than to their contract form, but if one is stuck with a given set of personalities delayed contracting (more so that contract type) can improve performance.
References
Wasserman, N. (2012). The founder’s dilemmas. Princeton University Press.
Moyer, M (2015).Slicing Pie: Fund Your Company Without Funds, Lake Shark Ventures LLC
1