This thesis aims to increase knowledge on firm performance with two separate objectives. The first analyzes how the measurement and analysis of supply chain performance could be enhanced. The second examines the connections between firm performance and supply chain practices.

Firm performance is considered to comprise both the financial performance of the firm, and intra-firm supply chain performance, which is further divided into cost performance in the form of logistics costs, service performance and asset utilization, as suggested by the research literature.

Based on extant literature, supply chain practices analyzed here include logistics outsourcing, supply chain collaboration, and information systems support. In addition, the role of firm characteristics such as size, manufactur­ing strategy and industry orientation is explored.

The thesis consists of four articles and an introductory part, which concludes on the results of the individual articles, presents a synthesis of key findings, and summarizes the managerial, methodological and theoretical implications of this research.

The research questions are addressed by analyzing firm-level empirical data from manufacturing firms operating in Finland. Data gathered from trading firms operating in Finland was used as a reference in one of the articles. The empirical material is derived from two sources: survey-based self-reported data from three national level Finland State of Logistics surveys from 2006, 2009 and 2010, and financial reporting data from responding firms for the corresponding years retrieved from official sources.

The data is analyzed with multiple analysis methods, including basic de­scriptive statistics, t-tests and analysis of variance and more advanced methods such as generalized linear models and generalized linear mixed models.  

The findings concerning the measurement and analysis of firm performance reveal that the performance measures are seldom normally distributed. Rather, skewness and kurtosis are common in the performance data. For instance, logistics costs, measured as share of turnover, consist mainly of small values, whereas also high relative costs occur, causing the distribution of logistics costs to be positively skewed. The data on financial performance, on the other hand, is concentrated close to the mean value, which is visible in the kurtosis of the data. The non-normality of the data should be taken into account when analyzing performance data, either by employing proper transformations of the data or by choosing analysis methods that are less dependent on assumptions of normality.

Analysis of the connections between chosen supply chain practices and firm performance revealed some interesting findings. First, most of the previous literature assumes a positive relationship between outsourcing and firm per­formance. That assumption was not supported in this analysis. It would seem that the effects of outsourcing are more dependent on firm level charac­teristics than on outsourcing itself. Firms are outsourcing their operations for many reasons, of which only some are directly associated with performance. Second, the results also indicate that the connection between outsourcing and perfor­mance is not linear, but could instead be curvilinear. In practice, this means that firms with either no outsourcing at all or with heavy outsourcing experi­ence better performance than firms that have not made an equally clear deci­sion in either direction.

A higher level of supply chain collaboration was found to be associated with lower logistics costs and better financial performance in the analyzed population. A higher level of information systems support was found to be associated with lower logistics costs, whereas the connection with financial performance was negative. Combined, the findings on supply chain collabora­tion and information systems support highlight the importance of collaborative actions within the supply chain. At the same time, they also warn about rely­ing too heavily on the technological capabilities of IT systems used to manage the supply chain.

This thesis provides contributions on: (i) managerial; (ii) theoretical; and (iii) methodological levels.

On the managerial level, the two main messages to practitioners are as follows: First, the connections between supply chain practices and firm per­formance are dependent on the firm’s characteristics, such as firm size, indus­try and the level of internationalization. With this in mind, rather than trusting readymade recipes for success, management should instead thoroughly analyze what the effects of different practices would be, considering the indi­vidual characteristics of their firm. Second, the results emphasize the key role of Supply Chain Management (SCM). Especially for firms that have expanded their supply chains beyond the domestic market, it is important to extend their perspective towards suppliers and customers in addition to service providers. Realizing this seems to be an essential factor in the success of the firm.

On a theoretical level, some of the results of this thesis are in line with the mainstream of existing literature, while at the same time some of the findings challenge the assumptions made in the literature. As anticipated, a higher level of supply chain collaboration was found to be positively associated with firm performance. The results on logistics outsourcing, on the other hand, question the previous assumptions about the positive connection with firm performance and instead highlight the firm level characteristics.

The thesis also contributes on the methodological level by providing addi­tional information on both the measurement and analysis of firm performance. The experiences of the research process support open-ended questions instead of pre-defined scales with pre-defined intervals as a tool to measure firm per­formance in a survey setup. Also, the results of the thesis highlight the non-normal nature of the performance data, suggesting tools for a more robust analysis. For example, some of the traditionally used research methods, such as ordinary regression analysis and structured equations modeling, are sensi­tive to the normality of the data. The findings of this thesis suggest the use of generalized linear models, which allows the use of non-normal distributions.

The empirical material of this thesis was obtained from national level surveys, with the respondent population covering most of the manufacturing industry operating in Finland. In that sense, the generalizability of the findings can be considered good, with certain limitations in mind; while the material covers well the manufacturing firms operating in Finland, the sample is limited to a single country. Internationally, some generalizations could be made, especially to manufacturing firms operating in similar small, export-oriented and developed economies. Micro-sized firms are omitted from the analysis of supply chain practices, which limits the generalizability towards micro-firms while simultaneously increasing it towards larger firms. In addi­tion, the findings concentrate on manufacturing firms and cannot therefore be directly applied to other main industries, such as the trading industry.