“A financial performance-based assessment of SMEs’ competitiveness – an analysis of Hungarian and US small businesses”

The study aims to identify the financial performance measures used as a proxy of the firm-level competitiveness dimensions of small and medium-sized enterprises and their competitiveness. By investigating the factors that affect competitiveness in gen- eral, those areas will be introduced, related to an identified competitiveness dimension. Financial and non-financial performance indicators will assess these areas. The paper considers competitiveness as an outcome variable, suggests a relationship between financial performance and the identified areas, and searches for the financial perfor- mance measures drivers. A panel data model was tested on Hungarian small and medium-sized enterprises (SMEs) and US SMEs. The collected data cover the period between 2013 and 2017. As a result of the applied panel regression, those variables were successfully identified that drive and could predict financial performance measures related to competitiveness. The research found a significant difference between the two-sample dataset results, which differences can be connected to country, industry, and, in general, to economic development characteristics. The results provide decision-making support and hint about the managerial tools and techniques aiming to control the firm characteristics, performance, and, eventually, firm-level competitiveness. Based on the results, further research can be dedicated to the development characteristics of firm-level competitiveness and the analysis of the relationship between the competitiveness dimensions and competitiveness itself.


INTRODUCTION
Competitiveness is a complex, multidimensional phenomenon, a relative concept that combines economic, managerial, political, historical, and cultural aspects (Waheeduzzaman & Ryans, 1996;Chaudhuri & Ray, 1997). Competitiveness in general and competitiveness analyzed at different levels (firm-level, sector level, regional focus of economic activities, national level) of understanding is debated among the academic society and practitioners and policymakers. At the macro level, country-specific competitiveness reflects a country's ability to manage its resources and competencies to increase the aggregate wealth of its citizens (Deakins & Freel, 2012). At the intermediate level, regional competitiveness is based on firms' combined competitive advantage and the comparative advantage of a local economy (Budd & Hirmis, 2004;Szerb & Ulbert, 2009). The roots of competitiveness can be traced back to the micro-level, as firms' economic value sums up the competitiveness of industries and, ultimately, nations. Firm-level competitiveness is the ability to deliver goods and services at the time and place, and in the form sought by buyers, while lowering the cost of production, providing superior value with superior returns on investment (Cook & Bredahl, 1991;Jiang, Jin, & Ren, 2017). From this research's point of view, the most appropriate definition for competitiveness comes from Basu (2011), who defines competitiveness as a multidimensional and complex concept that attempts to capture the process of fit between the firm and its dynamically evolving environment.
Small and medium-sized enterprises (SMEs) are rarely investigated through the lens of competitiveness. Depperu and Cerrato (2005) set out the economic and market performance indicators generally connected to competitiveness. The publicly available information about the frequently used performance measures, whether financial or non-financial, is limited and makes the research focus more larger-firm oriented. However, an SME is not a scaled-down version of large enterprises (Wisenthige & Guoping, 2016). Even though an increasing number of international studies focus on SMEs' competitiveness, there is still a lack of research focusing on Hungarian firm-level competitiveness. In contrast, neither Hungarian nor international focus aims to connect SMEs' competitiveness and financial or non-financial performance measures.
SMEs account for more than 25 million registered entities in the Old Continent and around 30 million in the United States, accounting for 99.8% and 99.4% of all enterprises. SMEs create the newest private-sector jobs and are responsible for 56.4% of the value-added generated by the business sector (EU Publication Office, 2020, p. 11). SMEs in Hungary are considered the central pillar of the economy, the primary source for jobs, and the economy's growth.
If an SME wants to become or remain competitive on the market, evaluating performance and effective resource utilization monitoring is inevitable. Competition is what requires SMEs to monitor their performance. Although performance measurement has potential benefits, it remains a too complex field for SMEs that requires too much effort from the firms, both in terms of human capital and financial resources.

LITERATURE REVIEW
Even though competitiveness has gained widespread attention in the economics and business literature over the past decades, the debate about the actual meaning of the concept of firm-level competitiveness is not yet solved, nor the issues about its assessment and measurement (Blandinieres, 2017). The various definitions available could be classified as competency-based and resource-based approaches. Competency approaches focus on the firms' abilities and capacity to fulfill the needs of their customers more efficiently than the competitors do (Budd & Hirmis, 2004); to use technology, quality, and performance for profit and growth (OECD, 1992); to compensate employees and provide incremental returns to shareholders (Buckley, Pass, & Prescott, 1988); to compete in a given business environment (Porter, 1990). Later, incorporating more dynamic aspects, a resource-based approach arrived with empha-sis on the specific characteristics of a given firm (Buckley et al., 1988). The competency approach identifies the crucial determinants of firms' competitiveness, such as strategy, structure, competencies, and capabilities, but now from an innovation, flexibility, adaptability, speed, and agility perspective (Hamel & Prahalad, 1989;Barney, 2001). Overall, firm-level competitiveness is usually related to the ability to fulfill firms' double purpose, make a profit through meeting your customer requirements, while continuously adapting, in short, to survive in the market (Chikán, 2008). The definitions mentioned above suggest that measuring a firm's competitiveness should incorporate quantitative measures and qualitative indicators.
From the above mentioned variable perspective derived (Chaudhuri & Ray, 1997), Depperu and Cerrato (2005) propose a classification, according to which competitiveness could either be considered as an input (driver) or output variable of firm performance (Akben-Selcuk, 2016). Buckley, Pass, and Prescott (1988) introduced a mixed concept, considering competitiveness as the driver of performance and a result of that. They distinguished competitive performance to assess the past and current firm performance; competitive potential to define the current and future competitive performance of a firm based on internal factors; firm capabilities to be able to translate competitive potential into actual or prospective performance. Building on that, in this research, competitiveness is approached through the dimensions of competitive performance, competitive potential, and firm-related capabilities.
Resources, may they be assets or labor-based resources that the company currently has, and those that it lacks and what it needs to grow, were the bedrocks of our models. The productivity of these assets determines competitive performance, while the resource of research and development (R&D) is the key to translating competitive potential, innovation into future performance. The level of innovation plays an important role in SMEs' business performance (Rostek, 2012).
Measuring company competitiveness is more like assessing it. There is not an appropriate proxy for firm-level competitiveness. The existing literature suggests approximating company competitiveness with some profitability, productivity, market performance measures, or unique indexes (Bhawsar & Chattopadhyay, 2015; Oral, Cinar, & Chabchoub, 1999; Liargovas & Skandalis, 2010). The paper connects competitive performance to productivity measures based on that well-documented and proven practice and resource-based logic. Figure 1 contains the applied regression models' dependent (total asset turnover, labor productivity, R&D expenditure) and independent variables (gross margin, size, age, growth, liquidity, leverage). These variables are meant to explain and predict the financial performance measures that drive the competitiveness dimensions.
Some would think measuring performance and measuring financial performance is an easy job, but it is far from being solved, especially at the SME level. With it being easy to understand results, financial performance measurement is a commonly recognized process to assess and ana-lyze firms; it provides essential information for investors, financial analysts, auditors, and management. Despite its advantages, as it was summed up by Katone (2016), SMEs rarely apply performance measurement systems due to the lack of capital and labor resources, and intense competition (Garengo et al., 2005); still, most SMEs focus more on financial indicators (Massalla, 1994;Monkhouse, 1995), relying mainly on accounting information and financial measurements (Carpinetti, Galdámez, & Gerolamo, 2008). This paper aims to give a clear picture of firm-level competitiveness and how SMEs should assess it from a financial performance-based perspective. It compares SMEs from a less developed country (Hungary) to SMEs from a developed country (USA) based on their financial performance-based competitiveness determinants.

METHOD
The following financial performance measures were picked for the research after breaking down competitiveness into competitive performance, competitive potential, and firm capability (see Fig ure 1): 1. Competitive performance dimension: • Total asset turnover, which ratio evaluates the efficiency of managing all the company's assets.
• Labor productivity corresponds to the total earnings before interest and taxes created by the company's labor.

Competitive potential dimension:
• R&D expenditure is the key indicator of the inputs into innovation, measured by the research and development costs related to net sales.
Mixed expectations can be connected to the explanatory variables. Since competitive performance is determined by market share, profitability, survival, growth, and productivity (see Figure  1), the most commonly used financial ratios were picked to express potential effects and predict power on the chosen measures in the financial analysis practice.
Market share's proxy was the firm size, which is frequently involved in the financial performance-based competitiveness analyses since it is one of the first and most influential decisions of a company. Führer and Michael (2004) argue that firm size affects a company's financial performance significantly and will increase more. Based on the traditional neoclassical view, one would anticipate a positive relationship between firm size and financial performance. Recent evidence shows firm size has positive and harmful effects on financial performance and, thus, overall competitiveness. Firms with larger size measured by their net sales, usually benefit from economies of scale, specialization, and diversification. They access capital and qualified human capital better and have higher negotiation power with stakeholders ( Growth is expected to positively affect financial performance; thus, competitiveness (Abor, 2005).
Regarding leverage, a positive relationship was documented by Ghosh (2000), Roden (1995), and Taub (1975), but according to Gleason, L. Mathur, and I. Mathur (2000) and Simerly-Li (2000), using more debt in the capital structures of the firm could lead to lower financial performance also.
The same mixed expectations could be defined regarding liquidity. Profitability and liquidity can complement each other, but excessive investments in current assets could end up in adverse effects   Table 1 summarizes the expected impact of the financial areas on financial performance, thus on competitiveness.

Financial performance measures Expected impact on financial performance/ competitiveness
Market share +/-Profitability Two sample countries were chosen to determine the effect, the explanatory power, and the predicting power of financial metrics on the competitiveness dimensions. SMEs' data from Hungary and the United States of America were collected from the Hungarian OPTEN database and the Wharton Compustat database running from January 2013 to December 2017. A shortlist of data was determined by filtering and using only those firms that have available data in all of the years investigated, and/or have most of the data that was needed for the research variables (since not every balance sheet and income statement data is obligatory to report about). In the case of the missing values, the linear interpolation method was applied based on company ID in the case of the Hungarian data set, on four-digit SIC codes in the US data set (industry bias is solved in the US sample). The final sample included 225 Hungarian firms per year and a total of 1,135 firm-years of observations and 982 US firms and a total of 4,607 firm-years of observations.
Based on identified competitiveness dimensions and the connected financial performance measures, the following explanatory variables' impact on firm-level competitiveness will be analyzed: , .
Financial performance f size age liquidity leverage gross margin growth This relation was estimated with dynamic panel data regression. Regarding the research questions and hypotheses, regression analysis is the most appropriate methodology for answering these questions. Because regression analysis is a process of analyzing the relationship between a metric dependent and one or more independent variables, it will enable us to construct an optimal model that will detect the firm-level competitiveness drivers. After all, in the regression calculation, the existence, direction, and strength of the relationship between the variables need to be estimated.  Based on the variables shown in Table 2, three different regression equations were assessed, each regressing one of the three dependent (TAT, LABOR, R&D) variables on the same set of explanatory variables. Due to the structure of the dataset, and to be able to deal with cross-sectional and time issues, panel data estimation was utilized. The following regression model was analyzed: where it Y was one of the dependant financial per-

RESULTS
The Hausman specification test's null hypothesis assumes that the preferred random model estimate is insignificantly different from the fixed effects estimate can be rejected (despite the relatively large chi-square values in the case of the US sample, since only the p-value is of interest).
Therefore, a fixed-effect panel data regression was applied. In addition to firm fixed effects, the yearly period fixed effects were added to the model based on the analysis conducted by Akben-Selcuk (2016), to take the influence of economic fluctuations also into account. The lagged values of the dependent variables were included as explanatory variables to assess the impact of past performance. To address potential heteroscedasticity concerns, the Eicke-Huber-White standard errors were included in the output tables (White, 1980).   Multicollinearity is tested with the absolute correlation coefficients, that need to be below the threshold value of 0.6, to avoid highly linearly related predictors. Table 4 shows the correlation matrix of the independent variables for the Hungarian data (a), and the US data (b), and proves that multicollinearity is not a concern in the research.
The following section presents the results of the panel regression analysis. It discusses the variables that significantly affect the financial performance based competitiveness of the Hungarian and US firms, assessed by their total asset turnover, labor productivity, and R&D expenditure. Table 5 presents descriptive statistics on the variables involved in the regression model. Since the natural log of all the data was calculated and used for further analysis, it is useless to interpret the table's mean values. Still, the standard deviation uncovers some crucial issues about the sample's characteristics. As it was expected, total asset turnover, employee number based labor productivity, and R&D expenditures vary the most among the companies involved. This standard deviation is much lower in the US sample. However, according to their assets, on average smaller companies landed in the sample (while filtering the available data, a filter was applied down to those companies that reported R&D related expenditures in the last five years in both the Hungarian and US dataset).
The results of the panel data regression are shown in Tables 6-8. According to the first regression model, in the case of the Hungarian companies, almost none of the variables employed proved to be significant in explaining the variation in financial performance measured by their asset-based productivity indicator, the total asset turnover. Only the lagged variable shows a significant relationship; thus, past productivity drives future productivity. The explanatory powers of the models reflected by the R-square results are both acceptable, in the case of the US sample, even higher than 40%.
In the US, sample size, growth, and past productivity positively and significantly affect productivity. A surprising result of what liquidity and the gross margin ratio is showing. The more liquid a company is, the less productive it becomes, and the same can be recognized regarding gross margin. The higher the profit ratio, the less productive the company will be. Although it needs to be further investigated, the phenomena could have to do something with the relatively high cash reserves of some innovative companies, even among small and medium-sized ones, and with the profit improvement strategy of the companies during the investigated period (cost reduction instead of revenue focus, which could lead to short term productivity concerns). The two sample's varying results mirror firm-specific, country-specific, and even size-specific effects.
The second regression model has even higher explanatory power than the first one when explaining the labor productivity ratio's variation in financial performance. The Hungarian companies show a significant negative connection with the profitability ratio among explanatory variables, and a positive connection to past labor productivity. The higher the profitability, the less productive the employees will be. Profitability shows a negative impact again, but this time on employee productivity. Does the higher profit come with or from less (productive) employees (a sign of automatization or motivation problems, employee treatment in general)? The US data also shows an expected significant and positive relationship with the profitability and growth rates. The remaining variables do not significantly affect labor productivity.
According to the third model, older Hungarian small and medium-sized companies are less willing to invest in research and development. This significant relationship cannot be found in the US dataset. The lagged variables drive the dependent variable again in both cases. In the Hungarian dataset size and in the US sample size, growth and liquidity are negatively related to R&D expenditures. While US companies that carry higher amounts of interest-bearing debt spend more on research and development larger (according to their asset amount), SMEs are less willing to innovate and spend money on R&D.   Figure 2 summarizes the relationship between the dependant and independent variables. Only four variables in the Hungarian sample ended up valid in the models, size, age, gross margin, and lagged values of each dependent variable. Liquidity, leverage, and growth rate cannot explain the productivity and innovativeness of the Hungarian SMEs. In the US data-based analysis, age is the only independent variable related to any dependent variable. Growth rate and lagged values have a significant impact on every dependent variable. This result highlights important financial areas and determi-nants for managers who seek more data and guidance about the variables affecting the firm's financial performance based competitiveness. Financial managers could build on these results and monitor these factors in their decision support system. At the same time, active control mechanisms could be directed on them for the sake of performance enhancement. Overall, profitability and size ended up as the factors that could improve (one way or another) financial performance and competitiveness (in both samples).

DISCUSSION
According to the research's empirical findings, financial and accounting data-based measures can For an investor in Hungary, "the smaller and younger, the better" SMEs with financial performance and competitiveness, and in the US, large, profitable SMEs with growth potential will be worth to invest.

CONCLUSION
Competitiveness has gained widespread attention in economic literature over the past decades.
This paper intended to authentically connect the competitiveness dimension to financial performance measures and indirectly to the latter's drivers. The financial performance was measured by the productivity and R&D expenditure related ratios as potential measures of one of the three competitiveness dimensions.
A firm's competitive performance was proxied by two performance/productivity indicators, by the total asset turnover and labor productivity, while competitive potential, the ability to innovate by the R&D expenditures as a percentage of net sales. Based on the research results, the following conclusions could be drawn. First, the two samples provided two entirely different pictures about the drivers of the Hungarian and US SMEs' financial performance. The Hungarian firms are still narrowing down their focus on profitability, incumbent firms are unwilling to innovate, and plans and predictions are built on historical parameter values.
In the US results, the gross margin ratio's profitability has a significant but mixed impact on the chosen dependent variables; in two cases, a negative relationship was documented (TAT, R&D). The leverage Table 9. Specific impact of the financial areas on competitiveness in the Hungarian and US samples Source: Own construction. variable has shown a positive sign for the R&D regression, suggesting that more leveraged US firms tend to spend more on research and development. The next surprising result that emerged from the empirical research is that the current ratio measure of liquidity has a negative impact on total asset turnover and R&D expenditures. As expected, a mixed result was found regarding the firm size. Firm size proved to positively and significantly affect asset productivity, but a negative one on R&D expenditures. Finally, growth has been found to positively impact productivity and a significant negative effect on R&D. In contrast, all three financial performance measures are positively related to their lagged values.

Financial performance measures Expected The impact in the Hungarian sample The impact in the US sample
The results derived from the analysis helped to explore the relationship between financial performance and the chosen firm-specific variables (size, age, liquidity, leverage, gross margin, growth, lagged values). The differences between the countries suggest economic development-related origins. Managers of an emerging country based SME could consider the Hungarian results, while managers of a developed country could consider the US results for decision support improvements. They could control some of the firm's resources and capabilities to obtain enhanced performance outcomes. Investors could also find useful hints regarding their investment aim and focus.
However, there are areas of this research that need to be further investigated. Further research can aim the search for country-specific or development-related characteristics of firm-level competitiveness. The relationship between the competitiveness dimensions and the chosen dependent variables needs to be tested as well.
Additional research potential is in the analysis of further emerging and developed country-based SMEs.