Designing human resource practices in multinational firms

Carlos Sánchez-Runde

Multinational firms face two basic dilemmas in the design of personnel practices. The first dilemma surrounds the process through which the practices are designed.
The second dilemma relates to the actual content of the practices

In order to be viable in free markets, companies must provide a minimally satisfactory solution to a series of basic problems. Two factors that directly affect the complexity of these problems are the organization's size and the diversity of environments in which it operates. Although - as in almost everything - there will be exceptions, these basic problems are particularly complex in multinational companies - large organizations operating in several countries.

In this paper, we will focus on three aspects of the basic problem of personnel management in large multinational companies. One aspect concerns the personnel practice formulation process and, particularly, the diverging pull between centralization and decentralization in the design of practices - in other words, whether practices should be designed from the corporate headquarters or in each subsidiary. The second aspect addresses the content of personnel practices and, particularly, the conflict between homogeneity and heterogeneity in the practices implemented. That is, to what extent practices should be the same in all the countries in which the multinational operates. Third, the interaction between process and content in the design of human resources practices and, particularly, the relationship between the level of centralization or decentralization in the design of practices, on the one hand, and the level of homogeneity or heterogeneity of practices, on the other hand. Throughout this paper, we will refer to the design of the practices, without directly discussing the implementation of that design.


Personnel practices:
The process of formulating centralization versus decentralization

A first important point to make is that talking simply of 'centralization' or 'decentralization' is mere academic licence. In actual fact, it is unlikely that there exists any case of pure centralization or decentralization; rather, there is a range of hybrid states between the two extremes.

Both centralizing and decentralizing the design of the system used for personal practices is associated with advantages and disadvantages. Without in any way claiming to be exhaustive, we will briefly discuss three aspects.

A first important point to make is that talking simply of "centralization"
or "decentralization" is mere academic licence. In actual fact, it is unlikely
that there exists any case of pure centralization or decentralization;
rather, there is a range of hybrid states between the two extremes

First, it is easier for the design of the human resource system to be controlled by a few people at the corporate headquarters than through several people in different countries. However, as several eyes in different places see more than the same eyes in the same place, decentralizing the system's design will tend to improve the quality of the final result.

Second, although designing the human resource practices system from the corporate headquarters will save overheads and the duplication of functions, something that would be difficult to avoid if each subsidiary were to be involved in the design, it is also true that duplicating functions may also duplicate the potential for the creativity and innovation required to obtain a high-quality design.

Finally, although decision-making may be quicker when personnel practices are designed entirely from a single location - corporate headquarters - the active involvement of the subsidiaries in the process will tend to make them more committed to the successful implementation of the system designed.

As we can see, the subject is far from simple. It can therefore be easily understood why - leaving to one side each alternative's greater or lesser rationality - multinationals end up being highly influenced by external fads and internal cycles. Generally speaking, and looking at what is most in vogue at the present moment, we would say that the pendulum is swinging towards centralizing operations as part of an effort to focus - among other things - on what the company does best and outsourcing accessory activities, reducing organizational inefficiencies as much as possible by downsizing, flatter organization structures and standardizing processes. However, on a practical level, each company is a world in itself and, applying the law of the pendulum, one can readily imagine that the future will favor increased centralization where until now decentralization has prevailed, and vice versa.


The content of Personnel practices: homogeneity and heterogeneity

Multinational companies must also decide the degree of similarity in the human resources practices implemented in their subsidiaries. The more homogenous the practices are, the simpler it will be, for example, to transfer personnel among subsidiaries. Likewise, increased practice homogeneity usually facilitates control and administration of the resulting system of practices.

On the other hand, the more one permits a certain degree of heterogeneity in personnel practices design, the better will be the fit between each subsidiary's individual features and the practices themselves. A variable, performance-linked compensation system, for example, may be indexed to each individual's performance or to that of the groups in which the work is organized. Linking the variable part of the compensation to individual performance will be more 'naturally' accepted, in principle, in individualistic societies - North American society, for example. However, in less individualistic societies - Japanese society, for example - it will be more 'naturally' accepted, in principle, to link the variable part to the work group's performance. In other words, once again, it appears that both homogeneity and heterogeneity in personnel practices have advantages and disadvantages.

And we say appears because some human resources experts insist on viewing this issue in a totally different light and argue that there is a group of personnel practices that are clearly more effective than the rest, irrespective of the industry in which the company competes or the country in which it operates. This body of practices are termed 'excellent' or 'universal' and, although their formulation varies slightly depending on who proposes it, they usually include the following: job security, extensive use of team work, rigorous selection, recruitment of high-level candidates, comparatively high salaries with a significant variable part, high investment in training, reduction in status differences - real and symbolic - among employees, extensive sharing of financial information with the personnel, and promotion from within. Of course, for those who share a vision of 'universal' personnel practices, the conflict between homogeneity and heterogeneity in human resources practices becomes meaningless: for them, any multinational company in its right mind will implement the set of 'excellent' practices in all of its subsidiaries, with the resulting homogeneity that this would produce.

Thinking about the process and the content together may help
us understand which alternatives would best suit each multinational

Formulating Personnel practices:
process and content

After having looked separately at some considerations on the process and content of the design of human resources practices, thinking about both aspects together may help us understand which alternatives - centralize or decentralize, homogenize or heterogenize - would best suit each multinational.

Leaving to one side the advantages and disadvantages we have mentioned when discussing the dilemma between centralizing or decentralizing the formulation of personnel practices, we will immediately see that there are two types of circumstances that advise in favor of a centralized design.

The first is when an approach based on what we have called 'universal' or 'excellent' practices is more viable. Contrary to what the followers of this approach maintain, it seems reasonable to think that there will be certain competitive situations in which these practices will be more or less meaningful. To put it more simply, many theorists accept that a system of 'excellent' practices is more adapted to companies that compete on the basis of frequent and intensive innovation, with customers who are relatively insensitive to the price of their products and services.

On the other hand, companies that compete on the basis of low prices and, rather than continuously developing new products, seek to defend relatively small market niches, would be advised to keep away from these 'excellent' practices. So the conflict will arise when a multinational company competes simultaneously on markets that require both types of practices.

Taking the situation to its extreme, one could then consider variants in practices for each of the businesses and/or countries in which the multinational competes. And as it is usual for multinational companies to operate in many different markets at the same time, the resulting system heterogeneity may become too complicated to handle. Having reached this point, it may be worthwhile simplifying design problems by imposing 'coffee for everyone' (a homogenous system) instead of 'drinks à la carte' (a heterogeneous system).

Following the same logic, decentralized designs will be more meaningful when the aim is to prevent sacrificing the fit between human resources practices and the way each of the multinational's businesses competes, or when the multinational company must be particularly sensitive to the specific cultural features of the countries in which it operates. This will lead to more heterogeneous personnel systems.

Decentralized designs will be more meaningful when the
multinational company must be particularly sensitive to the
specific cultural features of the countries in which it operates

Obviously, it will also be necessary to take into account other variables, such as each subsidiary's capacities at a given time. Irrespective of business or country, it is known that recently created subsidiaries that are still in the process of acquiring qualified personnel will need more help from the corporate headquarters when designing the human resource system. This will initially lead to more centralized and, no doubt, homogenizing designs. However, with time, as the subsidiary increases its skills and capabilities, more decentralizing and, perhaps, heteroge-neous alternatives could be considered.

Of course, as with every-thing else, in the final analysis, we must adapt to each company's specific circumstances. In this way, we can find a valid solution to the dilemmas that arise - and which we have discussed here - regarding the process and content of the formulation of personnel practices.

REVISTA DE ANTIGUOS
© IESE 1998