* This paper investigates whether a structural model explaining export performance of small- and medium-sized firms displays stability in a national context different from the one it was developed in.
* A comparative study between US and Austrian firms shows that the model developed in a small open economy does not produce satisfying results in a business environment which relies on a large home market.
* Given the differing factors influencing successful business activities abroad, this study calls for a more sensitive and considerate transfer of international marketing and management research results from one cultural context to another.
The last two decades have seen a considerable number of studies which deal with the identification of variables influencing a company's export success. Synoptic views are provided in the works of Bilkey (1978), Cavusgil and Nevin (1981), Miesenbock (1988), Gemunden (1991), Li and Cavusgil (1991), Douglas and Craig (1992) or Leonidou et al. (1998) for example. In particular, Miesenbock's analysis of state-of-the-art export success factor research and Gemunden's metaanalysis give a many-faceted overview of different research objects (e.g., companies, products), different success factors and their operationalization.
However, only surprisingly few studies in this research area rely on a crossnational data base. Gemunden (1991) identified a negligible number of research pieces which were based on more than one national sample (e.g., Burton/Schlegelmilch 1987, Dichtl/Koglmayr/Muller 1990, Dichtl et al. 1984, Hirsch 1971, McDougall/Stening 1975, Muller 1991, Rosson/Ford 1982), while Aaby and Slater (1989) listed only three cross-national studies in their literature review. From a methodological perspective, this lack of multi-country studies is highly questionable. Given the fact that the business environment is usually shaped by specific local influences such as physical, economic, political, legal, technological or socio-cultural factors, these local characteristics create a particular background for research conducted in this area and will most likely affect results in a typical way. Thus, findings derived in one national setting need to be carefully scrutinized for their transferability to other countries (Punnett/Ricks 1997, Terpstra/Sarathy 1997). Therefore, it appears highly recommended to re-assess the applicability of research findings that relate to a specific country to other regions as well (Hubbard/Armstrong 1994, Hubbard/Vetter 1996, Sharon/Amir 1988).
Reasons for the lack of relevant cross-national studies may be found in the extremely high demands that such work imposes on the researchers involved. As a rule, financial expenses are considerably higher and methodological problems are fundamentally more complex than in an intra-national context (Mayer 1978, Nasif et al. 1991).
At the same time, cross-national research offers numerous advantages over conventional one-country studies: as mentioned before, it lends itself particularly well to provide explanations with a wider range, i.e. beyond a single cultural context. Cross-national research may also take variables of the cultural background into account against which managers make their export-centered decisions (Bhagat/McQuaid 1982). As comparative research offers more variety and different degrees of intensity and expressions in behavioral patterns, the underlying conditions for hypothesis testing are more favorable (Kline 1988). Finally, cross-national comparisons assist in revealing differences and detecting and explaining behavioral patterns that do not exist in the researchers' home culture (Brislin/ Lonner/Thorndike 1973, Verma/Mallick 1988).
In line with these considerations, the objective of this paper is to investigate the cross-national generalizability of a well-tested explanatory model of export performance (Holzmuller/Kasper 1991, Holzmuller/Stottinger 1996) beyond the geographical region, where it was developed. …