Governance of Knowledge‐Intensive Firms in the Modern Market Economy

Abstract

We model contracting for joint production between workers and shareholders when investment in knowledge is non‐verifiable and the resulting specific human capital embedded in the workers is non‐tradable. The model explains how the effective cost of human capital services will vary depending on whether the investment in knowledge is financed by the workers or by the shareholders. We apply the results of the modeling to identify which firms are expected to gain and which to lose from posted trends in higher employability and lower empowerment of workers in modern market economies. Finally, we present conditions on the self‐interest of current shareholders to empower workers as a way to stimulate their investment in firm‐specific human capital.

Key Words:

• Corporate Governance
• Specific Human Capital
• Stakeholders
• Knowledge Management

JEL classifications:

• D23
• D86
• L14
• G34
• M55

Notes

1. The economics of R&D investment has documented that knowledge spillover to competing firms limits the appropriation of the returns from the investment for those firms that pay for it (Cassiman and Veugelers, 2002; Dosi, 1998; Griliches, 1979). In this paper, the spillover would affect equally the knowledge financed by workers and that financed by the shareholders, so it is not a relevant issue.

2. Efficient production requires that workers have the appropriate human capital, that is, the capabilities for the job they are assigned to, and also that they provide the appropriate effort when applying these capabilities. We focus on the first problem of human capital accumulation, and ignore the additional problem of motivating effort.

3. Antras (2003) also measures the relative technological intensity of a production input in terms of the elasticity of output with respect to the input in the production function. The justification for this can be found in equation (4) where, for given input prices, the intensity of use of inputs in the optimal production mix increases with the elasticity. The exposition could be extended to the relative intensity of use of knowledge versus physical capital, rather than two kinds of knowledge, but the interpretation of the results follows from what we present in the paper.

4. For a firm that initially insources the innovations because it is more cost efficient, a change in the parameters may induce the firm to shift from insourcing to outsourcing the innovations, if the conditions of the corollary above hold in the new environment. It can be shown that the cost saving from outsourcing, c(i) − c(e) increases with the initial level of the bargaining parameter α. Therefore, as λ is sufficiently high to justify the outsourcing, the firms that benefit the most from the decision are those where workers have higher bargaining power.

5. The interval will be non empty for values of the parameters that satisfy the condition α ≥ (1 − λ)/(2 − λ), which is the same condition that recommends that innovative workers finance the investment in knowledge, a condition that is assumed to hold in this part of the paper.

6. Cappeli (2008) observes that, today, firms have difficulty retaining talented managers whose training and career development the company paid for in the past. In response to this situation, he proposes that managers should pay a higher share of training costs than the one they paid in the past. Our analysis would support this recommendation, if training costs are measurable and contractible.

7. This conclusion may change if asset specificity is, in turn, a source of competitive advantage for the firm, and the value of the product increases as knowledge is more firm specific. Now, there may be a trade‐off between higher employability and more incentives to workers financing the investment in knowledge and lower revenues from selling the product.

8. For example, shareholders empower workers to make credible the promise that they will not change the production conditions in the future, so that the employability of workers would be lower than it was at the start of the relationship (Roberts and Van den Steen, 2000).

9. The respective pay‐off functions for workers in the two cases are , for the complete contract, and for the incomplete contract. The term λ(1 − α)H is the pay‐off that comes from outside opportunities; it will be zero when assets are fully specific, λ = 0, and if workers own the firm, α = 1.

10. Notice that we assume a world of certainty, where the risk attitude of the workers is irrelevant for the choice of the optimal contract. Under uncertain output and pay‐offs, the analysis should allow for risk‐aversion of workers in all solutions that imply risk sharing with the shareholders. In general uncertainty, risk‐averse workers and risk‐neutral shareholders will give the optimal second best contract where shareholders have more control and cash flow rights than in the case of no uncertainty or risk‐neutrality of workers.