trust toward the national bureaucrat to be an "honest bribee." that is, to deliver services as agreed after having taken the bribe, as another type of particularized trust which may seal corrupt deals. An important restriction for performing a corrupt transaction is the anticipation of the contracting party's opportunistic behavior (Williamson, 1975) and the costs resulting from it. Research argues that opportunism is more likely to occur in corrupt transactions than in legal ones. Several reasons are identified for supporting this. First, the public bureaucrat may renege on the corrupt deal when the offered bribe is worth less than the reputational gain from its denunciation (Della Porta & Vanucci, 1999, p. 195). Second, the risk of opportunism is high because corrupt deals take place outside the law and are not legally enforceable (Husted, 1994, p. 20; Lambsdorff, 2002b, p. 227; Rose-Ackerman, 1999, p. 92). Third, the public bureaucrat may create situations where he can force the businessperson to pay higher bribes than initially agreed upon due to the latter's high sunk costs. If corrupt deals are not legally enforceable, then alternative methods of assuring compliance must be designed (Rose-Ackerman, 1999, p. 96). Put differently: how can the costs of the potential opportunism be reduced in corrupt transactions after the bribe has been paid? We argue that this can be done either by engaging in social networks (based, e g.. on kinship or friendship), as mentioned in a previous section, or by drawing on trust. That is, the particularized trust between the bribe-payer and bribe-receiver may serve as an informal enforcement mechanism of the corrupt deal. The higher the trust toward the public bureaucrat, the higher the predictability of the corrupt deal, and thus the higher the likelihood of engaging in corruption. These considerations lead us to the following hypothesis: Hypothesis 6: The higher the reputation of the public bureaucrat to stick to the deal after having taken a bribe, the higher the entrepreneur's likelihood of engaging in corruption. Methods Data and Variables The database employed in this study is "The World Business Environment Survey (WBES) 2000." a survey conducted by the World Bank Group to measure the impact of the political, economic, and legal institutional make-up on the firm performance. We use a sample based on 2.576 firm responses from 20 transition and mature market economics. The WBES questions have been employed and validated by different scholars (for recent studies see, e.g., Uhlenbruck et al., 2006). Additional information on the dataset can be accessed at http:/info.worldbank.org/governance/wbes/index2.html. For a descriptive overview of corruption and its antecedents, 20 countries are classified into different regions. Among the transition economies, Russia (n = 470) and less investi- gated economics from the Trans-Caucasus (Armenia, Azerbaijan, Georgia; n = 350) and Central Asia (Kyrgyzstan, Kazakhstan, Uzbekistan; n = 334) are selected to compare them with more developed and better studied countries from Central-Eastern Europe (Czech Republic, Hungary. Poland, Slovakia, Slovenia, Estonia, and Lithuania; n = 871). Indus- trialized states from Western Europe (Germany. United Kingdom, France. Sweden; n = 363) and North America (United States and Canada; n = 201) are used as a comparator group to transition countries. The respective WBES questions for the dependent and independent variables as well as their measurement scales are displayed in Table I.\f