【正文】 ude banks in very small transition economies, ., the Baltic countries and Slovenia, and those in less advanced transition economies that have only recently restructured the banking system, ., the former Soviet Union, Albania and the other Balkan states. In the next section, we present a brief description of the privatization experiences in these six countries to establish that the strategies and the timing of privatizations are sufficiently different to allow us to use these experiences as the basis for an empirical analysis of privatization. Section 3 describes our dataset and presents the results of testing for differences in means across bank types for several measures of bank performance and for several bank characteristics. Section 4 characterizes briefly our methodology of deriving profit and cost efficiency measures from stochastic frontier estimates that allow for country and year effects directly in a pooled data set. In this section, we relate the bank efficiency scores, as well as a measure of financial performance, to the type of ownership and the method of privatization in secondstage regressions. Section 5 concludes with a brief summary focusing on policy implications. 2. Bank Privatization in Six Transition Economies Pretransition banking sectors were designed to meet the needs of a centrally planned economy(CPE).Intermediation between savers and borrowers was internalized within the state banking apparatus basically through a system of directed credits to stateowned enterprises for both investment needs and budget allocations for the working capital necessary to meet the output plan. In most CPEs, large specialty banks performed specific functions. A state savings bank, with an extensive branch network, collected virtually all household deposits. A foreign trade bank handled all transactions involving foreign currency. An agricultural bank provided shortterm financing to the agricultural sector. on bank funded longterm capital projects and infrastructure ,banking activities were both subservient to the plan and segmented along functional lines in CPEs. In the transition economies (TEs), the first step in banking sector reform involved creating a twotier system with mercial banking activities carved out of the old central bank. At the beginning of the decade, the new banking sectors in the former CPEs consisted of the newly created mercial banks and the specialty banks, both types having universal banking licenses, along with a few foreign Greenfield banks and often many relatively undercapitalized Renovo domestic private banks that were born under lax entry banks had virtual monopolies in their core activities, ., the savings bank was often the only entity with an extensive enough branch network throughout the country to collect primary deposits. Typically, three or four large banks dominated the emerging banking sector in a TE. Both the newly created mercial entities and the specialty banks were stateowned initially. Hence, structural segmentation, a proliferation of weak small domestic private banks, and stateownership of the large banks were the major features of banking sectors in TEs at the beginning of the 1990s.These legacies affected the banking sectors in all of the countries in our sample with the exception of Croatia, which was part of Yugoslavia. From the 1950s, mercial banks in Croatia as well as the other republics were not stateowned but were owned collectively according to the Yugoslavian system of selfmanagement. Virtually all foreign exchange deposits collected by the republiclevel banks were remitted to the National Bank of Yugoslavia in Belgrade in exchange for credits in dinars. Upon succession in June 1991, the Yugoslavian government froze the foreign exchange deposits of Croati