صداقت مدیریت و سرمایه گذاری خارجی در کشورهای در حال توسعه

The possibility of facing dishonest local managers is an important factor explaining foreigners’ hesitations about investing in developing countries. The first part of the paper analyses the optimal decision rule of a manager who is able to transfer into his own hands a part of the output of the firm. It is shown that in a two-period framework with incomplete information about the nature of the manager, a rational expectations equilibrium exists where managers’ behaviour and investors’ expectations are mutually consistent. In particular, some young managers may aim at building a reputation of an honest person, then behave dishonestly when getting old. The global performance of an economy hosting a large number of managers is investigated in the second part of the paper, where analysis is cast in an overlapping generation framework. The frequency of dishonest managers, development prospects and global profitability appear to depend on the duration of the relationship between investors and managers.

Recent trends show that foreign direct investment, wherein a foreign company either builds from scratch or acquires a controlling stake in an existing facility, has become a major method of financing the developing world (UNCTAD, 1999). Many factors are considered as traditionally determining the foreign direct investment flow. All foreign investors have the same concerns: political stability, economic openness, ready access to inputs at reasonable prices and laws and regulations that are fairly and transparently enforced. Investors also look to the size and growth of domestic markets and closeness of major international markets; cheap raw materials and labour may also be important assets.† Unfortunately, in many countries from Asia, Africa, Latin America and Eastern Europe, ambiguity in definition of property rights, widespread corruption of state officials‡ and inefficient legal systems pave the way of local managers to appropriate a share of the firm’s wealth without being subject to legal sanctions. This paper develops a formal model on the lines of classical studies on policy credibility (e.g. Barro and Gordon, 1983; Backus and Driffill, 1985; Barro, 1986; Vickers, 1986) to analyse how the risk of facing dishonest managers influences foreign investment in a small developing country. It puts forward the incidence of reputation concerns on managers’ individual and collective behaviour and highlights the impact of the length of the relationship between investors and managers on the global performance of such an economy. In the first part of the text, we focus on the optimal decision of a representative manager with a two-period decision horizon (he is first young, then old). To introduce fraud risk in a simple way, we assume that each manager has the choice between diverting a positive share of the firm’s income or preserving the firm’s integrity. The representative manager wants to maximize his expected money income, made up of a statutory wage proportional to the firm’s turnover and the diverted funds. Managers differ with respect to the subjective value they attach to $1 of diverted money; it may be nil for the ‘‘white knight’’ manager, or may be very large for the dishonest one. The foreign investor does not know the nature of the manager he faces, but knows the statistical distribution of this characteristic in the population of managers. It will be shown that a reputation effect occurs when investors agree to keep † Substantial literature analyses the role of foreign direct investment in developing countries and its determining factors. See Donges and Wieners (1994), Dunning (1995) or the surveys of Agarwal (1980), Rayome and Backer (1995), Meyer (1998). ‡ Literature on the economic consequences of state officials’ corruption is surveyed by Bardhan (1997) or Tanzi (1998).

The possibility of facing dishonest managers is an important factor in explaining foreigners’ hesitations about investing in developing countries. This paper investigates the influence of reputation concerns on managers’ optimal decision whether to cheat or not, then uses the main microeconomic results to analyse the global performances of the hypothetical economy. In the first part of the text we have analysed the joint decision of representative pair manager–investor. Information is incomplete as investors do not know the priorities of the manager they face. It is shown that a rational expectations equilibrium with Bayesia learning does exist. The properties of this equilibrium depend on the duration of the relationship between managers and investors, itself related to the duration of the firm. If investors agree to keep employed a honest young manager, but to fire a dishonest one, some managers may perform honestly during the first period of the 248 D. BESANCENOT AND R. VRANCEANU contract only to manipulate investor’s expectations, then steal during the last period. However, such behaviour would contribute to reduce the global number of dishonest managers in the economy. If investors undertake only short-term projects, manager’s incentive to build a reputation of an honest person vanishes. As in this approach the frequencies of facing one of these managers are endogenous, several macroeconomic implications can be neatly inferred. On this purpose, the two-period model is recast in an overlapping generations framework, which brings the economywide population of managers into the picture. At any time period, a generation of young managers coexists with a generation of old managers. Aggregate capital and profits are obtained by adding up individual amounts. The model allows to portray two extreme cases. In a country where investors enter only short-run relationships, the reputation effect is ruled out and actual manager misbehaviour is, indeed, high. The global profitability of this economy is rather low and investment and output are meagre. In a country where investors undertake long-term projects, more young managerswill behave honestly. Our results suggest that there is an inverse relationship between the frequency of dishonest mangers in a developing country and the global inflow of foreign investment. Given that profitability and growth are larger in the ‘‘high’’ equilibrium, this would justify external support for long-run projects. In a closely related field of research, several empirical studies found that foreign investment is adversely affected by official corruption. For instance, Mauro (1995, 1996) has shown that corruption, as measured by a standard corruption index, has a significant negative impact on the ratio of investment to GDP. According to his estimate, a 2% reduction in the corruption index could raise the growth rate by 0
5%, via a positive impact on investment. Wei (2000) analysed the impact of corruption and other relevant variables on bilateral stocks of FDI between twelve source countries and forty-five host countries in 1993. He found that a 1% increase in corruption index may be associated to a reduction in the stock of inward FDI by as much as 26%. Using firm-level FDI data from Eastern Europe and the CIS countries, Smarzynska and Wei (2000) showed that the probability of investing in a given country is negatively correlated with state officials corruption and that, in high corruption countries, foreign investors tend to favour more joint ventures than wholly owned subsidiaries. Finally, Tanzi and Davoodi (1997) presented some evidence that high corruption tends to be associated with low operation and maintenance expenditures and poor quality of infrastructure.† † Other references to empirical analyses that put forward a negative relationship between corruption and growth can be find in surveys by Aron (2000) and Tanzi and Davoodi (2000). MANAGER HONESTY AND FOREIGN INVESTMENT 249 In general, the authors of these empirical studies adopt the conventional point of view according to which corruption is determining investment or growth. Indeed, in their regression models, the corruption index is one of the right-hand side independent variables. Our paper contributes to this literature by showing that manager economy-wide corruption and the amount of investment are jointly determined. It also suggests that both variables depend on the nature of the investment projects. But as a matter of principle, we can only agree with these authors in claiming that efficient support to development should start by creating institutions that prevent dishonest behaviour from spreading: well-defined property rights, awell-functioning juridical system and a transparent and accountable government.