PRODUCTIVITY IN SERVICE, MANUFACTURING, AND HYBRID INDUSTRIES: AN EMPIRICAL STUDY.
Productivity is one of the most fundamental concepts in economics and is important for various reasons. Growth in productivity determines the growth of the economy, and particularly the labor productivity largely determines the level of well-being of the working class. Measuring changes in productivity and understanding relative differences in productivity across various sectors of the economy, industries, and even on a firm-level is an important work that allows us to get an insight into how the economy is evolving over time and where the source of growth lies.
One of the important features of this evolution, in almost all developed and developing economies, is that the service sector has expanded steadily in the recent decades. In 2017, according to the World Bank, service sector based on value added accounted for about 65 percent of the global GDP. For the United States this number is reported to be at about 77 percent (The World Bank, 2018, a). The better we understand productivity in the service sector of the economy the better grasp we will have on the overall economy.
The share of the economy generated by the service sector has steadily increased at the expense of the manufacturing sector. Table1 shows the growing gap between the sectors is starker for the developed economies. With increased globalization and heightened levels of international trade flows, manufacturing industries in developed economies are facing stronger competition from those in developing countries where the cost of labor is cheaper. As such, the developed economies' reliance on service sector has increased at a higher rate. Furthermore, according to the World Bank, the employment share of services (modeled ILO estimate) in the United States increased from 72.1% in 1997 to 78.9% in 2017 (The World Bank, 2018, b).
According to the Bureau of Labor Statistics (BLS), productivity has grown faster in the manufacturing sector as compared to the nonfarm business sector in the past three out of four decades (U.S. Bureau of Labor Statistics, 2018). Such trend is likely to emerge when service sector productivity is growing slower than manufacturing productivity. This trend appears to be troubling because of the increasing reliance of the economy on the service sector in terms of employment and output.
While there have been some attempts to compare productivity across sectors, the evidence is sparse. Most studies in this field discuss the issues of measuring productivity in services but do not attempt to systematically compare productivity across sectors (Dean & Sherwood, 1994; Nachum, 1999; Biege, Schroter, & Weissenberger-Eibl, 2011). The sector-level estimates of productivity seem to suggest that service firms have lower labor productivity than their manufacturing counterparts. While this observation might be due to the fact that manufacturing firms have long started to substitute labor with other factors of production, capital and technology, while the service industry is highly dependent on labor, we propose an alternative view.
This study contributes to the literature by comparing labor productivity across firms that have been categorized not only based on the standard industrial classifications (SIC) or the North American Industry Classification (NAICS), but also based on the textual analysis of business descriptions. It was observed that there are firms within the manufacturing sector (by NAICS code) that also produce and bundle services with their manufactured goods. Similarly, there are many firms classified in the service sector (by NAICS code), which also manufacture goods. Therefore, current estimates of labor productivity for manufacturing and service sectors can be further disentangled.
The novelty of this study is that we develop a new classification of sector segmentation based on both NAICS and the textual analysis of business descriptions together. Based on textual search of business descriptions, we divide firms into three broad categories:
(1) purely manufacturing firms,
(2) purely services firms,
(3) hybrid firms.
Hybrid firms themselves are divided into two groups. The first group includes firms that are classified as manufacturing firms based on the NAICS code but seem to also offer services as suggested by their business description (servitized manufacturers). The second group consists of companies that belong to the service sector based on the NAICS code, but also manufacture and sell goods (manufacturized service). With this new classification, the authors of the current study measure labor productivity of firms in each category and arrive at new conclusions regarding the productivity differences between industries.
First, this study shows that pure manufacturing firms are indeed more productive than pure service firms. Second, servitized manufacturers are more productive than pure manufacturing firms and manufacturized service firms are more productive than pure service firms. That is, hybrid firms are more productive than firms that only produce goods or firms that only offer services. Third, the higher productivity of the hybrid firms goes hand in hand with higher levels of profit margin and return on equity.
Increasing reliance of the economy on the service sector is a global phenomenon but it more severely affects the developed nations. Consequently, these economies are becoming more dependent on services in terms of generating output and providing employment opportunities. In a concurrent trend, manufacturing firms are also increasingly relying on the provision of services as a source of revenue. Such trend is referred to as servitization and there is significant amount of research that focuses on servitization, servitization trends, and the impact of servitization (Vandermerve & Rada, 1988; Neely, 2007; Neely, 2008; Baines et. al., 2009).
As per manufacturing productivity, Katayama, Lu, and Tybout (2009) pointed out that "firm-level data on physical quantities of output, capital, and intermediate inputs are typically unobservable" and conventionally one has to use various proxies to measure productivity (e.g. real sales revenue can be used as a proxy for real output). The authors developed an alternative approach to capture productivity and showed that their estimated (welfare-based) productivity indices were only weakly correlated with conventional productivity measures.
Measuring productivity of the service firms has even more complexities. Blois (1984) listed some of the issues of measuring productivity in general and for service companies in particular. For service companies, the main challenges are due to the intangible nature of both inputs and outputs (i.e. components that cannot be measured or are extremely hard to capture). Another aspect of services that makes it hard to measure productivity is 'the lack of storability' when there is no way for workers to utilize their unused capacity later.
The lack of storability makes service productivity dependent on the variability of demand. Dean and Sherwood (1994) provided a great summary of the problem of measuring the output and productivity of many service-oriented industries by pointing to two main difficulties: the role of consumers in the production of a service as well as the challenge to disentangle the service that is a part of a bundle.
In relation to measuring labor input, Dean and Kunze (1992) claimed that data problems are more pronounced in the service sector. Most nonfarm self-employed and family workers work in service industries and the data on hours worked (a proxy for labor input) for those two categories of workers is very sparse. Siegel and Griliches (1992) categorized problems of measuring output in service industries into three broad categories:
(1) the lack of data on service industries (i.e. data on output and prices),
(2) the lack of clarity about what is being transacted (e.g. it is unclear what services correspond to the payments made to their providers. Gallouj and Savona (2009) referred to this problem as "the immateriality of the outcome of production and delivery"),
(3) the issue of quality change.
Another strand of research, in addition to discussing measurement issues in services, focuses on the productivity gap between manufacturing and services. Nachum (1999) postulated that the commonly held belief that productivity in services lags behind compared to productivity in manufacturing might stem from the inadequacy of existing data or measurement issues. Using a sample of Swedish management consulting firms, he showed that his measure of productivity in services (as compared to a manufacturing-based measure) correlates better with other measures of firms' performance.
Tello (2017) primarily focused on the impact of innovation on productivity in the service sector. That study used data on Peruvian firms and, constrained by the kind of data available, utilized real value added per worker as a simplified measure of productivity in services. The value-added data came from sales data and the value-added estimates were obtained by using the industry level average ratios of value added relative to value of production obtained from input output tables. For the purposes of comparison, Tello analyzed the same relationship for the companies in the manufacturing sector. The estimates of labor productivity in that study suggested that service sector firms are on average less productive relative to manufacturing firms.
Sample and Data Collection
To construct the novel industry classification, the authors of this study used quarterly data from the proprietary COMPUSTAT database. The data spans from 1965 to 2015 and includes all publicly traded U.S. firms. Using the Bureau of Labor Statistics classification, we identified manufacturing firms as those belonging to 2-digit NAICS codes 31, 32, and 33, and service-providing firms as those belonging to 2-digit NAICS codes 42, 44, 45, 48, 49, 51, 53, 54, 55, 56, 61, 62, 71, 72, 81. Following the previous literature, the firms providing utilities and financial services were omitted from the study analysis due to the fact that they are subjected to peculiar government regulation.
Measurement of Variables
The authors then searched the business description of each firm for clues of the scope of the firm's operations and constructed the main variables of interest. If a business, which is categorized as a manufacturing firm according to its NAICS code, uses the words "service" or "solutions" in its business description, then it was taken as an indication of this firm offering some form of a service along with its manufactured product. Then this business was coded as a servitized manufacturing firm. For perspective, examples of servitized manufacturing firms include Computer Identics Corp and Engineered Support Services. The following excerpts are from their respective business descriptions:
"Computer Identics Corp designs, manufactures, markets and services standard bar code products and systems for the data collection and material handling/industrial markets; and also offers a variety of support services." "Engineered Support Systems, Inc. designs, manufactures, and supplies integrated military electronics, support equipment, and technical and logistics services. It operates in two segments, Support Systems and Support Services."
Similarly, if a business, which is categorized as a service-providing firm according to its NAICS code, uses the words "produce" or "manufacture" in its business description, then we took it as an indication of this firm offering a manufactured product along with its services. We then coded this business as a manufacturized service firm. Examples of manufacturized service firms include CompuDyne Corporation and International Game Technology. The following excerpts are from their respective business descriptions:
"CompuDyne Corporation [...] manufactures and integrates physical and electronic security equipment and services for local, state, and federal government agencies." "International Game Technology designs, develops, manufactures, and markets casino-style gaming equipment, system technology, and game content for land-based and online markets worldwide. The company operates in two segments, North America and International."
The union of all servitized manufacturing and manufacturized service firms is called hybrid firms. If a firm, which is categorized as a manufacturing firm according to its NAICS code is not identified as a servitized manufacturing firm as a result of our textual evaluation, then this firm is defined as a pure manufacturing firm. Similarly, if a firm, which is categorized as a service firm according to its NAICS code is not identified as a manufacturized service firm, then this firm is defined as a pure service firm.
In the final panel data set, there are 172,596 firm-year observations belonging to 14,841 unique firms. 75,669 of these firm-year observations correspond to 5,935 pure manufacturing firms, while 65,006 of the firm-year observations correspond to 6,517 pure service firms. Moreover, 18,587 of these firm-year observations correspond to 1,187 servitized manufacturing firms and 13,334 of the firm-year observations correspond to 1,202 manufacturized service firms, together forming 31,921 firm-year observations of 2,389 hybrid firms.
All of the firm financial data that were used in the construction of the variables also came from the COMPUSTAT database. In the spirit of prior literature, the authors constructed the dependent variable--labor productivity--as the ratio of the firm level quarterly revenue to the number of the firm's employees and they took the natural logarithm of this ratio. As a measure of the firm's size, the authors used the natural logarithm of its fixed assets. In order to be able to control for the capital intensity per employee in some regression specifications, the authors also constructed a variable by dividing a firm's fixed assets by the number of its employees and taking its natural logarithm.
In order to investigate the differences in labor productivity across firms in different sectors, the following model was employed in the spirit of Bernard and Jensen (1999):
[InLP.sub.i] = [[beta].sub.0] + [[beta].sub.1][Sector.sub.i] + [[beta].sub.2]Employeesi + [[beta].sub.3][Employeesf.sub.i] + [[beta].sub.4][lnCapital_Stock.sub.i] + [[beta].sub.5][lnCapital_Stock.sub.i] + [[beta].sub.6]lnCIR + [a.sub.Year] + [[epsilon].sub.i] (1)
where i denotes firms and j denotes the type of the firm's sector (depending on the regression specification, either belonging to pure service sector, manufacturized service sector or servitized manufacturing sector), InLP is the natural logarithm of labor productivity, Sector is a dummy variable denoting the type of the firm's sector and is the main variable of interest, Employees is the count of the number of employees (in thousands), Employees (2) is the square of the number of employees in order to control for the possibility that the number of employees might have a concave relationship with labor productivity, lnCapital_Stock is the logarithm of firm's capital stock measured by its fixed assets (gross value of machinery, equipment and plant after depreciation), lnCapital_Stock (2) is the square of the capital stock variable in order to control for the possibility that the size of the capital stock might have a concave relationship with labor productivity, lnCIR is the natural logarithm of the capital intensity ratio. [[alpha].sub.Yea]r is the year fixed effect, and e is the error term.
The specification in equation 1 follows the literature on estimating labor productivity differences. Bernard and Jensen (1999) used a similar specification in order to estimate the labor productivity premium of exporters relative to non-exporters. Similarly, this specification was used to estimate the labor productivity of firms in one sector relative to firms in another sector.
The initial analysis of productivity differential between pure service and pure manufacturing firms confirms that the former group is significantly less productive. In this setup, productivity premium is defined as percentage difference in labor productivity between pure service firms and pure manufacturing firms. Table 2 shows the results of the regression analyses where the main variable of interest is Pure Service Firm, which takes the value of 1 if the firm offers only services without bundling them with a manufactured good, and the value of 0 if the firm only manufactures goods without bundling them with a service. In all the specifications of the regression function, pure service firms appear to have both economically and statistically significant productivity disadvantage.
To be precise, when comparing a pure service firm to a pure manufacturing firm, both with the same number of employees and similar size of fixed assets, the pure service firm, on average, appears to generate less revenue per employee. Since the dependent variable of the regression function in all the specifications is taken at the natural logarithm, the coefficient of the variable of interest can be interpreted as a percentage difference in productivity. In the strictest specification, where the quadratic forms of the Employees and Log of Capital Stock variables, as well as the year and industry fixed-effects are included, pure service firms are, on average, about 7% less productive compared to pure manufacturing firms.
More interestingly, there is a significant productivity variation within the types of service firms themselves. When the service firms are dissected further and separated as purely service offering firms and service firms that also offer manufactured goods (i.e. manufacturized service firms), the manufacturized service firms are significantly more productive. Table 3 shows the results of these analyses. The main variable of interest here is Manufacturized Service Firm, which takes on the value of 1 if a firm identified by its NAICS code as a service firm also manufactures a product as judged by the textual analysis of its business description, and the value of 0 if a firm identified by its NAICS code as a service firm does not manufacture a product and hence is a pure manufacturing firm. In column 4, in the strictest specification of the regression form, the hybridized version of the service firms (i.e. manufacturized service firms) are approximately 17% more productive than their pure service offering counterparts.
A similar trend emerges when the manufacturing firms are separated into two groups: those that only offer manufactured goods and those that produce manufactured goods and also bundle them with some sort of service (i.e. servitized manufacturing firms). Just as the hybrid service firms are more productive than pure service firms, the hybrid manufacturing firms (i.e. servitized manufacturing firms) are also more productive than their pure manufacturing counterparts.
Table 4 shows the results of these analyses where the main variable of interest is Servitized Manufacturing Firm, which takes on the value of 1 if a firm identified by its NAICS code as a manufacturing firm also offers some service as judged by the textual analysis of its business description, and the value of 0 if a firm identified by its NAICS code as a manufacturing firm does not offer any service and hence is a pure service firm. In column 4, in the strictest specification of the regression form, the hybridized version of the manufacturing firms (i.e. servitized manufacturing firms) are approximately 18% more productive than their pure manufacturing counterparts.
Overall, the hybrid firms are significantly more productive than their pure manufacturing or pure service-offering competitors. This productivity differential lends itself to a more beneficial cost saving structure, such that the hybrid firms also exhibit greater profitability.
Panel A of Table 5 shows the mean values of profit margin for hybrid firms, pure firms (pure manufacturing and pure service firms combined), and the results of the t-test for the difference of profit margins between the hybrid and pure firms.
The profit margin variable has been winsorized at the 2% and the 98% levels in order to decrease the distorting pull of outliers in the averages. The t-test shows that hybrid firms have, on average, a greater profit margin at a statistically significant level.
The higher level of productivity and profit margin of hybrid firms also implies a better "bang for the buck" for the shareholders. For every dollar of equity that the owners invest in an average hybrid firm, they receive a higher return compared to pure firms.
Panel B of table 5 shows the mean levels of ROE for hybrid and pure firms as well as the results of the t-test for the difference in ROE between the hybrid and pure firms. The ROE variable has been winsorized at the 2% and the 98% levels in order to decrease the distorting pull of outliers in the averages. Not surprisingly, the hybrid firms earn their shareholders a higher return on every dollar of equity invested.
Productivity is an essential element of a business unit and of the economy at large. When measured at highly aggregated levels, it may not accurately reveal the most productive forces of the economic activity. In the face of heightened foreign competition, especially in the manufacturing field, American firms have had to differentiate their products from their pure manufacturing or pure service competitors elsewhere in order to carve themselves a specialized niche and accrue a loyal customer base. Therefore, hybrid form of doing business where a manufacturing firm bundles its goods with some form of a service; or a service firm also starts offering a manufactured good, has become an important distinguishing phenomenon.
This study proposes a way of disaggregating the manufacturing sector into servitized and pure manufacturing sectors; and the service sector into manufacturized and pure service sectors on the basis of textual analysis of the 10-K business descriptions of individual firms. Servitized manufacturing and manufacturized service firms together form the set of hybrid firms.
The investigation of the productivity variation across these sectors indicates that while standalone pure service firms are less productive than standalone pure manufacturing firms, hybrid firms are more productive than their counterparts in both of these categories. Most interestingly, among all firms that are involved in manufacturing, those that also bundle their goods with services (hybrid firms) are on average 18% more productive than pure manufacturing firms, and among all service firms, those that also manufacture and sell goods (hybrid firms) are on average 17% more productive than pure service firms.
The results indicate that the most productive firms are those with lines of business that allow an additional dimension to differentiate in the product markets. By bundling a manufactured good with a service, or vice versa, hybrid firms are able to offer more valuable products, and therefore capture larger revenues per employee and generate larger profits for each dollar of sales, as evident by the variation in profit margins. Moreover, hybrid firms also appear to be better investments for the shareholders as they yield higher levels of ROE compared to non-hybrid firms.
Baines, T. S., Lightfoot, H. W., Benedettini, O., & Kay, J. M. (2009). The servitization of manufacturing: A review of literature and reflection on future challenges. Journal of Manufacturing Technology Management, 20(5), 547-567.
Bernard, A. B., & Jensen, J. B. (1999). Exceptional exporter performance: cause, effect, or both?. Journal of International Economics, 47(1), 1-25.
Biege, S., Schroter, M., & Weissenberger-Eibl, M. (2011). Product adaptation for industrial product-service systems--characteristics, motives and challenges. In Functional Thinking for Value Creation (pp. 33-38). Springer, Berlin, Heidelberg.
Blois, K. J. (1984). Productivity and effectiveness in service firms. The Service Industries Journal, 4(3), 49-60.
Dean, E. R., & Kunze, K. (1992). Productivity measurement in service industries. In Output measurement in the service sectors (pp. 73-107). University of Chicago Press.
Dean, E. R., & Sherwood, M. K. (1994). Manufacturing costs, productivity, and competitiveness, 1979-93. Monthly Lab. Rev., 117, 3.
Gallouj, F., & Savona, M. (2009). Innovation in services: a review of the debate and a research agenda. Journal of Evolutionary Economics, 19(2), 149.
Katayama, H., Lu, S., & Tybout, J. R. (2009). Firm-level productivity studies: illusions and a solution. International Journal of Industrial Organization, 27(3), 403-413.
Nachum, L. (1999). Measurement of productivity of professional services: Anillustration on Swedish management consulting firms. International Journal of Operations & Production Management, 19(9), 922-950.
Neely, A. (Ed.). (2007). Business performance measurement: Unifying theory and integrating practice. Cambridge University Press.
Neely, A. (2008). Exploring the financial consequences of the servitization of manufacturing. Operations Management Research, 1(2), 103-118.
Syverson, C. (2011). What determines productivity? Journal of Economic literature, 49(2), 326-65.
Siegel, D., & Griliches, Z. (1992). Purchased services, outsourcing, computers, and productivity in manufacturing. In Output measurement in the service sectors (pp. 429-460). University of Chicago Press.
Tello, M. D. (2017). Innovation and productivity in services and manufacturing firms: the case of Peru. CEPAL Review.
The World Bank. (2018, a). World Bank national accounts data, and OECD National Accounts data files. Retrieved March 5, 2019, from https://data.worldbank.org/indicator/NV.SRV.TOTL.ZS.
The World Bank. (2018, b). International Labour Organization, ILOSTAT database. Retrieved March 5, 2019, from https://data.worldbank.org/indicator/SL.SRV.EMPL.ZS.
U.S. Bureau of Labor Statistics. (2018). Labor Productivity and Costs. Retrieved March 5, 2019, from https://www.bls.gov/lpc/prodybar.htm.
Vandermerwe, S., & Rada, J. (1988). Servitization of business: adding value by adding services. European Management Journal, 6(4), 314-324.
University of the Pacific
William Jewell College
About the Authors:
Vusal Eminli is an Assistant Professor of Finance at the Eberhardt School of Business at the University of the Pacific. He received his Ph.D. and M.S. degrees in Economics with concentrations in Finance and Industrial Organization from Purdue University, and his B.A. degree in Mathematics and Economics from Berea College. Dr. Eminli's primary research interests are in corporate finance, corporate governance, and industrial organization.
Yuriy Bots is an Associate Professor of Economics in the Business and Leadership Department at William Jewell College. Dr. Bots received his B.A. and M.S. degree in Economics in Ukraine where he worked for two years as an economist. He continued his education at Purdue University and received his M.S. and Ph.D. in Economics with concentration in International Trade. Dr. Bots' current research is focused on the importance of service firms in international trade and leadership issues in the non-profit organizations.
Table 1 The Share of the Service Sector in GDP Country Services, value added (% of GDP) Percentage point change 1997 2017 United States 71.8 77.4 5.6 High-income countries 65.0 69.9 4.9 Low- and middle-income countries 49.6 53.8 4.2 Source: World Bank, https://data.worldbank.org/indicator/NV.SRV.TOTL.ZS Table 2 Pure Service Firm Productivity Premium over Pure Manufacturing Firms (1) (2) VARIABLES Log Labor Log Labor Productivity Productivity Pure Service Firm -0.0291 (***) -0.0900 (***) (0.109) (0.107) Employees -5.60e-05 -0.000567 (***) (7.33e-05) (6.57e-05) [Employees (2) Log Capital Intensity Ratio 0.495 (***) 0.328 (***) (0.00283) (0.00278) Log Capital Stock -0.00109 -0.0854 (***) (0.00202) (0.00194) Log Capital [Stock (2) Constant 2.801 (***) 2.331 (***) (0.0144) (0.0245) Number of Firm-Year Observations 135,130 135,130 Number of Firms 12,426 12,426 Year FE NO YES (3) (4) VARIABLES Log Labor Log Labor Productivity Productivity Pure Service Firm -0.146 (***) -0.0711 (***) (0.109) (0.106) Employees -0.00303 (***) -0.00221 (***) (0.000140) (0.000128) [Employees (2) 1.20e-06 (***) 9.38e-07 (***) (7.92e-08) (7.19e-08) Log Capital Intensity Ratio 0.479 (***) 0.324 (***) (0.00286) (0.00282) Log Capital Stock -0.125 (***) -0.109 (***) (0.00296) (0.00272) Log Capital [Stock (2) 0.0179 (***) 0.00449 (***) (0.000303) (0.000290) Constant 2.970 (***) 2.387 (***) (0.0147) (0.0247) Number of Firm-Year Observations 135,130 135,130 Number of Firms 12,426 12,426 Year FE NO YES Standard errors in parentheses (***) p<0.01, (**) p<0.05, (*) p<0.1 Table 3 Manufacturized Service Firm Productivity Premium over Pure Services Firms (1) (2) VARIABLES Log Labor Log Labor Productivity Productivity Manufacturized Service Firm 0.227 (***) 0.171 (***) (0.0550) (0.0533) Employees 0.000298 (***) -0.000299 (***) (7.32e-05) (6.58e-05) Employees (2) Log Capital Intensity Ratio 0.456 (***) 0.366 (***) (0.00363) (0.00340) Log Capital Stock -0.0176 (***) -0.122 (***) (0.00251) (0.00248) Log Capital Stock (2) Constant 2.985 (***) 2.260 (***) (0.0154) (0.0380) Number of Firm-Year Observations 67,589 67,589 Number of Firms 7,038 7,038 Industry FE YES YES Year FE NO YES (3) (4) VARIABLES Log Labor Log Labor Productivity Productivity Manufacturized Service Firm 0.221 (***) 0.172 (***) (0.0550) (0.0533) Employees -0.00165 (***) -0.00185 (***) (0.000153) (0.000139) Employees (2) 5.84e-07 (***) 7.70e-07 (*** (8.16e-08) (7.42e-08) Log Capital Intensity Ratio 0.454 (***) 0.364 (***) (0.00363) (0.00345) Log Capital Stock -0.157 (***) -0.163 (***) (0.00398) (0.00366) Log Capital Stock (2) 0.0182 (***) 0.00665 (***) (0.000406) (0.000388) Constant 3.159 (***) 2.343 (***) (0.0158) (0.0382) Number of Firm-Year Observations 67,589 67,589 Number of Firms 7,038 7,038 Industry FE YES YES Year FE NO YES Standard errors in parentheses (***) p<0.01, (**) p<0.05, (*) p<0.1 Table 4 Servitized Manufacturing Firm Productivity Premium over Pure Manufacturing Firms (1) (2) VARIABLES Log Labor Log Labor Productivity Productivity Servitized Manufacturing Firm 0.222 (***) 0.179 (***) (0.0288) (0.0280) Employees -0.00115 (***) -0.00164 (***) (0.000171) (0.000153) Employees (2) Log Capital Intensity Ratio 0.513 (***) 0.252 (***) (0.00381) (0.00390) Log Capital Stock 0.0149 (***) -0.0380 (***) (0.00279) (0.00258) Log Capital Stock (2) Constant 2.690 (***) 2.437 (***) (0.0151) (0.0261) Number of Firm-Year Observations 88,778 88,778 Number of Firms 6,856 6,856 Industry FE YES YES Year FE NO YES (3) (4) VARIABLES Log Labor Log Labor Productivity Productivity Servitized Manufacturing Firm 0.236 (***) 0.182 (***) (0.0286) (0.0279) Employees -0.00762 (***) -0.00391 (***) (0.000295) (0.000268) Employees (2) 9.16e-06 (***) 4.71e-06 (*** (6.18e-07) (5.59e-07) Log Capital Intensity Ratio 0.472 (***) 0.247 (***) (0.00389) (0.00394) Log Capital Stock -0.102 (***) -0.0553 (***) (0.00366) (0.00337) Log Capital Stock (2) 0.0195 (***) 0.00376 (***) (0.000381) (0.000365) Constant 2.919 (***) 2.489 (***) (0.0157) (0.0265) Number of Firm-Year Observations 88,778 88,778 Number of Firms 6,856 6,856 Industry FE YES YES Year FE NO YES Standard errors in parentheses (***) p<0.01, (**) p<0.05, (*) p<0.1 Table 5 Comparison of the Profit Margins and the ROE between Hybrid and Pure Firms Panel A. Comparison of the Profit Margins of Hybrid and Pure Firms (Manufacturing and Service Combined) Hybrid Firms Pure Firms Hybrid--Pure (difference) Mean -0.1477 -0.2561 0.1084 (***) p-value 0.0000 Standard Error 0.0048 0.0031 No. of Firm-Year Obs. 31658 138329 Panel B. Comparison of the ROE of Hybrid and Pure Firms (Manufacturing and Service Combined) Hybrid Firms Pure Firms Hybrid--Pure (difference) Mean -0.0224 -0.0427 0.0203 (***) p-value 0.0000 Standard Error 0.0031 0.0016 No. of Firm-Year Obs. 31727 139783
|Printer friendly Cite/link Email Feedback|
|Author:||Eminli, Vusal; Bots, Yuriy|
|Publication:||International Journal of Business and Economics Perspectives (IJBEP)|
|Date:||Sep 22, 2019|
|Previous Article:||MULTI-GROUP R&D AND SALES PERFORMANCE IN THE UNITED STATES.|
|Next Article:||BANK INTEREST RATE PASS-THROUGH IN THE EURO AREA: THE IMPACT OF THE FINANCIAL CRISIS.|