The Governance of Human Capital: Lessons from Emerging Industries

THE GOVERNANCE OF HUMAN CAPITAL: LESSONS FROM EMERGING INDUSTRIES

(‘Organizzazione e governance del capitale umano: lezioni dalla nuova economia’)

Anna Grandori

Alessandro Usai

Luca Solari

Silvia Bagdadli

CRORA

Istituto di Economia Aziendale

Università Bocconi

viale Isonzo 23

20135 Milano

Italy

Ph.+39.02.5836 2633/2637

Fax +39.02.5836 2634
e-mail:

This paper is concerned with the governance of human capital as distinct from the governance of the employment relation. Most traditions of analysis in the area of the organization of work – in economics and sociology alike – have been concerned with how the provision of work can be coordinated with the provision of the services rendered by other types of resources – technical, financial, physical. However, while for technical and financial resources, the difference between a relation of resource investment and a relation of service provision is usually clear, the distinction has been quite blurred as far as human relations are concerned.

In general three units of analysis can and should be distinguished : the level of services rendered by the use of resources, the level of the ‘bundles of resources’ from which they stem, the level of actors - physical or juridical persons – possessing the resources and/or delivering the service (Penrose 1959, Grandori 2001). They should not be collapsed into one another, in principle, as different combinations between actors, resource ownership and acting (service delivering) can be envisaged.

Management literature talks indifferently of ‘human resource management’ and of ‘labor relations’, and seldom distinguishes between the level of ‘human resources’ (knowledge and competence ) and the level of the actors possessing them. ‘Human resources’ are often considered as a synonym of ‘people’, thereby neglecting the issue of people’s strategies about the allocation and use of their resources.

Sociological analyses have been overly concerned with labor as a service (strange to say, given that Marx – often quoted in that literature – was one of the first to notice that there is a difference between selling labor and selling ‘labor-power’; or perhaps understandably because that tradition of thought also takes for granted that ‘capital’ and the ‘means of production’ are separated from work).

Organizational economic analysis is concerned with ownership of assets, and does distinguish between labor transactions involving or not involving investments in specific assets (Williamson 1979). However, both TCE and agency theory remain focused on the governance of ‘transactions’ – exchange of labor services across technologically separable interfaces – and do not examine fully the governance of human capital investments in technologically unseparable combinations among themselves and/or with other assets. In addition, their analyses are conducted under quite restrictive assumptions on the configurations that human capital providers’ preferences may assume - randomly distributed in Williamson’s analysis (1981); risk-averse and effort averse in agency theory (Levinthal 1988) .

Overall, the governance of labor relations when the relation is an exchange of services is therefore well analyzed. It gave rise to typologies of ‘contracts’ effective for the purpose: ‘market-like’, ‘hierarchical’, ‘obligational’ and ‘relational’ (Williamson 1979, 1981) (Figure 1). They differ by time horizon and by the mix of governance mechanisms employed: respectively, the matters regulated by the contract, include only prices and service specification, or also and mainly supervision, rules and procedures, norms of reciprocity and fair conduct. These governance arrangements can be fairly well predicted by features of the labor transaction, such as task interdependence and the observability of performances, even without hypothesizing any specific investment of human assets.[1]

Interdependence

- +

+ market-like contracts obligational contracts

Difficulty in performance

evaluation

- hierarchical contracts relational contracts

Figure 1. The governance of labor transactions: a typology of exchange contracts

Source: Adapted from Williamson (1981)

A further array of contracts, however could be prospected for labor relations characterized by specific human capital investments rather than just by service transfers – which are analysed in less detail. In this case the situation would be similar to that of an investor of specific technical capital. The relevant questions for an investor are: (Shleifer, Vishny 1996): what governance arrangements can induce to invest? What the guarantee are that the investment will be fully compensated? What the prospects are to be able to collect the capital back, how much accrued, at what time? The relevant governance mechanisms for solving those problems typically include ‘property rights’ and not only the organizational rights mainly considered in labor exchange contracts (rights of action, information, communication, decision and control). Property rights, in the broad sense currently admitted also in organizational economics, include the ownership of assets, the rights to residual rewards from their use, residual decision and control rights on how the assets are used, the right to sell the asset or the services deriving from it ( Fama , Jensen 1983a,1983b). Contracts including a specification of how those rights are allocated can be defined ‘associational’ contracts, as distinguished from ‘exchange’ contracts, as the formers establish a continued association among dedicated assets, expose them solidarly to risk, and entitle the providers of those assets to sharing the residual economic results of the activity (Galgano I.II.III 1974)[2].

A typology of associational contracts. A first general hypothesis, therefore, is that human capital investments are to be regulated by associational contracts (Becker 2000, Hart and Moore 1986, Aghion and Tirole 1994).

More refined hypotheses should however be advanced in order to describe and explain the variety of associational contracts actually employed[3]. They involve differences in the extent to which different property rights are shared by human capital providers. The following governance mechanisms can be usefully distinguished to capture those differences:

-  ownership of firm assets, in the form of own capital or shareholding;

-  rights on residual rewards, in the form of pay for performance or of stock options;

-  non residual, risk-free rent sharing provisions, in the form of above average base income;

-  decision and control rights on how the ‘professionally conducted organized economic activity for the production or exchange of goods or services’ constituting the firm should in fact be conducted;

-  residual decision and control rights on rewards allocation and on human capital mobility;

-  ‘beautiful exit’ rights (golden parachutes, time horizon of stock option rights).

Supposing that specific investment of human assets are made, we should hypothesize that a non contingent feature of the governance package is that relevant decision and control rights on activities are allocated to those principals/agents. They would care a lot about what particular activities the firm will undertake, as their own capital will go together. They would not be 'quasi-indifferent' on the nature of tasks, as it was hypothesized to occur in a classic ‘employment relations’ (Simon 1955), as one reason to make a human specific investment (and actually to explain why such risky undiversified investments are made at all) may well be the interest in performing particular tasks. The effective combinations of those mechanisms are predictable on the basis of some additional variables, that qualify the types of specific human capital investments made. Relevant candidate variables are the measurability of performance and the riskyness of activities.

Current conceptualizations about corporate governance structures, and the relations between financial capital providers and human capital providers have focused on the motivational problem of providing ‘powerful incentives’ to those in charge of ‘control’, in order to orient the latter’s (unobservable) behaviors in the interests of the former actors. This frame neglect the risk problem as well as the problem that the latter actors make specific investments in the firm as well; to the extent they do so, they are in a position of ‘principals’ as well.

The riskier an activity is, the more results are influenced by exogenous variance, the less intense the optimal incentive is, ceteris paribus (Milgrom, Roberts 1992). Hence we should hypothesize that the incidence of pay for performance, should not be extremely high in the governance of human capital investments in risky activities. Unmeasurability of outcomes is also an impediment to pay for performance. We refer in particular to unmeasurability due to the intrinsic complexity of activities (uncertain categories of consequences, long term effects, poor quantifyability) rather than just to team production effects, which could be managed by group contingent rewards. By contrast, individual and group pay for performance should be diffused wherever activities stay in known causal relations with measurable outcomes.

What mechanisms can substitute for high powered incentives at the individual and group level, in conditions of low measurability of partial outcomes? There, shared property rights over firm assets and activities, globally intended, are in order (Fama, Jensen 1983a, 1983b). Figure 3 hypothesizes that ownership will not be separated from control in condition of limited risk and highly complex, non measurable outcomes; while financial investments will be more separated from human capital investments in highly risky ventures, where otherwise, human capital providers, which can not diversify easily their human investments, would bear excessive risk. Then, the governance of human capital investment relations in highly risky activities, should be characterized by more diversified investors (internal and external), a higher incidence of ‘premium’ base income and of ‘golden parachutes’, a higher incidence of stock options (a relative low risk and collective form of contingent reward) or analogous ‘bonuses’ linked to positive variance in firm’s results in non listed firms.

A typology of human capital ‘markets’. A second array of hypotheses on the governance of human capital investment relations, concerns the modes and channels through which these managerial resources are attracted, evaluated and selected, in emerging industries.

Figure 2. The governance of specific human capital investments: A typology of associational contracts

How is managerial Human Capital "traded" in the labor market? What are the mechanisms governing the exchange of managerial Human Capital across organizational boundaries, in emerging industries?

In the last decades it appeared clear how the internal and external labor market partition was unable to explain important phenomena. As it happened with regard to other important economic phenomena, the internal/external labor markets divide seems to be insufficient to explain Human Capital mobility. Granovetter (1974, p. 26) already noted how "several factors mitigate against perfect labor markets. Inertia as well as social and institutional pressures ... (but) the factor most relevant to the present discussion is imperfect information". Years later, Roth and Xing added that "markets may require a good deal of organization. This runs counter to the view implicit in much of the economic literature, which is that markets are largely self-organizing" (1994: 2). It is difficult for managers to possess complete information about job opportunities, job characteristics, job attractiveness and on the other hand it is often difficult for companies to get access and subsequently assess the real value of the managerial Human Capital in the labor market. According to the economic view candidates and employers try to overcome the limitation of information adopting active "search" behaviors, maximization of utility by rational actors using marginal principles pervades these models. Perhaps however the way job-search theory has most adversely affected research has been in deflecting attention from the large number of jobs that, by almost any definition, are not found by search[4] (Granovetter, 1995: 142). In the last decades however, starting from the founding work of Granovetter (1974), scholars, mainly coming from sociology, have started to study the social side of labor markets and transactions, looking at the hidden governance structures laying beyond the "market" idealtype. Granovetter (1974) found that 55,7% of workers interviewed found their job through personal contacts, others (Corcoran, Datcher and Duncan, 1980:12), following 5.000 American families through a panel design, found that 52% of white men and 47,1% of white women found their jobs through friends and relatives. These evidences have been confirmed in the following years by studies on workers of manufacturing plants (51,4% of jobs found through personal contacts; Marsden and Campbell, 1990: 68) and on young American people employment (at least 40% of 17-25 year-old workers found jobs through personal contacts; Staiger, 1990: 7). The "social governance" of job markets has been confirmed outside the US. In Britain during the 1970's and 1980's between 30 and 40% of respondents to a national survey found jobs through friends and relatives (Fevre, 1989). Watanabe's (1987) study of 2003 male workers in the Tokyo metropolitan area, shows that of those who change jobs, 54,6% do so through personal contacts. Boxman, DeGraaf and Flap (1991), in their survey of 1.359 Dutch top managers, found that 61% had found their job through contacts. Finally in a recent survey conducted among alumni of an important Italian University of Business studies, it emerged that "only" 23,3% of the respondents declared to have found their current job through personal contacts[5] (Santoro, 2000), however, the percentage was higher for respondents in the highest age and income categories.

In short, it seems that not only Social Capital may be crucial in the creation of Human Capital (Coleman, 1988) but that it could be even more important as a governance mechanism in the transfer and exchange of it.

One limitation of large part of the mentioned research has been a lack of interest in discriminating between effective and efficient forms of social networks and ‘clanistic’ and ‘oversocialized’ ones. To the purposes of understanding how human capital investments are regulated in emerging and innovative industries, for example, it would be crucial to distinguish between effective forms of professional network – responding to the difficulties in assessing human resource potential – and ‘socially over-embedded’ networks generating ‘local search biases’, thereby reducing innovation potential. Building on research focusing on high level work and innovative industries, as well as on relevant theory, we are therefore going to construct a typology of effective forms of network governance that takes into account some relevant contingencies, which parallels what has been done for contractual forms.

For what it concerns managerial networks, it has been showed that managers have larger networks than non-managers, that managers' social networks (core discussion networks) tend to include a greater number of professional contacts, and that managers interacting with co-workers tend to have higher job rewards (Carroll and Teo, 1996). It has also been argued that managers with well established networks have better career potential (Jackall, 1988; Krackhardt, 1990). Finally it seems that managers, because of the long hours they spend under stress at the workplace, tend to achieve higher intimacy with co-workers, transforming apparently professional ties into especially "close relationships" (Carroll and Teo, 1996). Most of the research however has been conducted within organizational boundaries and not so much is known about the role and the characteristics played by managers' social networks when managers change jobs.