1. Miracle or Myth?

Firms that fall into this category are typically small and mid-size firms that, among other things, share specific governance, finance, human resource relations, and long-term orientation (Audretsch et al. 2018). They tend to be family-owned and usually sell specialized machineries and components for the B2B market (The Economist 2014). This differentiates Mittelstand from the broad concept of small and medium-sized enterprises (SME) that only considers firm size criteria.

According to Simon (2012), who coined the term hidden champion in the early 1990s, they are defined as firms that, in general, generate below 5 billion Euros in revenues, belong to the top three in their respective market globally or are market leading on their continent, and are relatively unknown to the broader public. Estimates of the number of HCs in Germany range from approximately 1,300 (Simon 2018) up to around 1,700 (Rammer & Spielkamp 2019).

The Eisenmann SE is an industry supplier with a diversified portfolio. According to company statements, the acquisition and transaction of major projects caused high losses in 2018. Due to decreasing revenues and ongoing losses, the family firm decided to file for insolvency in July 2019 to promote a strategic realignment of the firm (FAZ 2019).

Additional support comes from Schmieder (2018) who states that, except in the automobile industry and the service sector, HCs showed significantly higher growth rates than the average German firm in the respective industry sector in 2014.

Findings by Schmieder (2018) support this observation and show similar vertical integration levels. Moreover, HCs show higher integration levels across all industries compared to the average German firm in the sample.

Empirical research by Rammer and Spielkamp (2019) shows that HCs put significantly more emphasis on employee creativity and individual responsibility compared to control group firms.

Rammer and Spielkamp (2019) state that 66.7% of HCs engage in continuous in-house R&D, compared to just 52.9% of control group firms.

Simon (2012) states that HCs on average invest 6.0% of the revenues in R&D which is double the amount of what German firms that engage in R&D show in general. According to Simon (2018), this is also significantly more than the 1,000 largest firms globally spend in relation to their revenue.

According to Simon (2012), HCs have costs of approx. 0.5 million Euros per patent (includes R&D-related costs) whereas large corporations spent 2.7 billion Euros per successful patent filing. The number of patents per employee is 31 for HCs and 6 for large corporation. Simon does not elaborate if the patents are based on internal R&D or are acquired externally.

The actual revenue limit was 1.5 billion Deutschmark (Simon 1996). Based on the official conversion rate, this  corresponds to roughly 766 million Euros. Simon (2009) raised the threshold to 3 billion Euros before he set the current limit to 5 billion Euros.

Venohr and Meyer (2007) investigate 200 HCs that are derived from Simon’s initial list of 450 HCs. Schlepphorst et al. (2016) use secondary data originally created to investigate fast growing firms. Schmieder (2018) bases his investigation in parts on a list of German world market leaders by Langenscheidt and Venohr (2015).

Freimark et al. (2018) compare HCs with SMEs and large corporations based on an average survey sample of 230 firms. Schlepphorst et al. (2016) observe 60 HC and 346 non-HCs.

The MIP is a yearly survey conducted by the Center for European Economic Research (ZEW), the Fraunhofer Institute for Systems and Innovation Research (ISI), and the Institute for Applied Social Sciences (infas). It is commissioned by the German Federal Ministry of Education and Research (BMBF) and gathers information on the innovation activities of German firms. The survey targets all firms with more than five employees in manufacturing and business-oriented services industries. It covers over 7,000 to 8,500 responding firms each year. The survey is based on a stratified random sample and, thus, allows for extrapolation of the survey results to the total firm population. Every two years the MIP is part of the CIS of the Statistical Office of the European Commission (EUROSTAT) which serves as the basis for the European Innovation Statistic (ZEW 2019). See, e.g., Rammer and Peters (2013) for details about the Mannheim Innovation Panel.

This view is supported by a comparison of the average HC size. Simon (2012) reports an average of 2,037 employees, whereas the extrapolated 1,800 HCs in Rammer and Spielkamp’s (2015) study show 731 employees on average.

See Appendix A.1.1 for the complete German Innovation Survey (2017- reference year 2016). This exemplary survey equals the surveys for the reference years 2012 and 2014 in all categories that are used for the identification strategy.

According to Rammer and Spielkamp (2015), significant market share is dependent on market volume. Therefore, if the overall market is below 200 million Euros, firms need to have at least 10% market shares. For market volumes between 200 and 500 million Euros the threshold is 7%, for market volumes between 500 and 1,000 million Euros the threshold reduces to 3% and markets larger than 1 billion Euros only require a 1% market share.

Rammer and Spielkamp (2015) argue that in order to show above-average growth rates, firm growth rates need to be at least 10% higher than the average growth rate of all firms in the respective industry.

Based on a screening of publications by Simon, 197 HCs (see Appendix A.1.3) are identified. From the initial list of HCs, 44 have taken part in the MIP in the reference years 2012, 2014 or 2016 at least once and are, consequently, used to verify the results of the identification strategy. The list of German market leaders was provided by Müller (2019). Based on an initial 524 firms in his list, 151 could be used in the matching process.

According to Simon (2012), global market leadership can be assumed when a firm is among the top 3 in its respective global market or market leader at the continental level.

One could argue that firms with a primary sales market outside of Germany should always report export shares of more than 50%. Yet, in the case of larger corporations, survey respondents often indicate export shares explicitly for the German entity and not for the entire corporation. Due to the introduced three factor approach, this bias should not affect the robustness of the identification of global players among firms in the MIP.

See Appendix A.1.2 for the Stata do-file for the identification process based on MIP data.

Only 12 of 319 firms failed this criterion. In most cases erroneous market share or export values caused the incorrect HC identification.

See Table A.3.2 in the appendix for the details of the matching results and an overview of the remaining firms at each identification step.

See Table A.3.1 in the appendix for a detailed breakdown of HCs and non-HCs for all survey years between 2012 and 2018.

There is no statistically significant difference between the groups of unclassified firms and classified firms in terms of revenues (p-value for equality of means = 0.32) or firm age (p-value for equality of means = 0.93). Moreover, there is no significant difference between the groups and industry affiliation, except that classified firms are 30% more likely to operate in the manufacturing of electronics and optical products (NACE 26). See Figures A.2.1, A.2.2 and A.2.3 in the appendix for a graphical density comparison between the two groups.

The extrapolation is based on extrapolation factors provided by the ZEW. Table A.3.3 in the appendix provides a detailed description of sampled and extrapolated HCs across different industries. Due to a response rate of 25 to 30%, the average weight per firm is about 25. However, there is great variation among sectors and size classes due to a disproportionate sampling structure (Rammer & Spielkamp 2019).

Table A.3.3 in the appendix provides the detailed distribution of HCs across industries.

For exact numbers see the innovation panel documentations of the respective year, available at https://​www.​zew.​de/​en/​publications.

The gross sample size reduces to 77,630 observations and 11,093 distinct firms after the HC classification procedure.

Due to panel imbalance, only 546 of the classified firms provide innovation data in all survey waves. Of these, 308 report zero revenues based on innovations for the entire period.

The variable for incremental innovation performance is derived through a subtraction of the radical innovation share from the overall share of revenues from innovation.

The productivity numbers in Euros refer to the exponentiated values of the mean and maximum logarithm of 11.446 and 12.737, respectively.

As an example, the survey for the reference year 2018 collected data on the profit margin for 2017 and 2018.

The HC effect for the logarithm of employee productivity is 0.356 in column (1b). Based on this result, the exponentiated value of the coefficient is 1.4276 which translates into a 42.76% higher employee productivity.

Compared to control group firms within similar age, size and industry groups, Rammer and Spielkamp (2019) report an 8.24 percentage points higher revenue growth rate for HCs. However, their identification strategy includes above industry average growth rates as a prerequisite to qualify as a HC.

Surveys suggest that compared to other industries, manufacturing related industries, in which most HCs operate, show mid-range profit margins (Destatis 2009).

Due to substantial decreases in observations, no controls for firm innovation activity are applied as robustness tests in this case. However, a test with decreased samples still shows significant, although slightly reduced, HC effect sizes.