As the Korean economy has entered a low-growth trend since 2010, the Korean government has established an innovative growth strategy with the goal of discovering growth engines that will lead the Korean economy. Arepresentative policy was to expandR&D investment. With this strategy, R&Dinvestment inKorea increased sharply, ranking second in the world in terms of R&D share of GDP in 2019. R&D investment led the growth of Korean economy through the accumulation of knowledge capital. However, as R&D’s contribution to the growth of the total factor productivity(TFP) is facing a downward trend, concerns about the decrease in its efficiency are growing. Meanwhile, the advent of the era of the fourth industrial revolution, digital transformation, and transition to the eco-friendly society, accelerated by the recent COVID-19, have increased the need for new investment strategies other than R&D investment. Although the importance of various innovative investment is fully recognized, Korea’s innovation investment is concentrated only on R&D investment. This can be explained by the low awareness of various innovative investments and the lack of research. As the definition and classification system for innovative investment have not been established, database and basic research for innovation investment are lacking. Based on these discussions, this study aims to categorize innovation investment and construct data for systematic analysis for accurate research on innovation investment and improvement of investment efficiency.

We defined and classified innovation investment, and constructed innovation investment data based on macro and micro statistics. In this study, innovation investment as classified into six categories according to the innovative activities of the Oslo Manual. (① R&D investment, ② marketing investment, ③ Intellectual property(IP) investment, ④ human resource investment, ⑤ software investment, ⑥ organization investment) The innovation investment data based on macro statistics were constructed by reclassifying CHS intangible asset, which has the similar classification system with the innovation investment types defined in this study. we constructed data by reclassifying variables related to CHS intangible assets established by KISDI. Meanwhile, micro data refers to firm-level data. The innovation investment data based on micro data were constructed using financial statements provided by KISVALUE and R&D cost information on the cost statement provided by KISLINE. The data was constructed by matching types of innovation investment and accounts in financial statements. Each type of investment is classified into the types included in the asset account, the types included in the expense account, and the type included in both the asset and expense account. software and IP investment are included in the asset account, so the process of converting assets into investment(flow) is required. Marketing and human resource investment are included in the expense account, so there is no need to convert them into flow data. R&D investment is included in the assets and expenses account, so the process of converting assets in flow data is necessary. After data construction, an asymmetric resistance fence method was used to detect outliers, and then firms with outliers were removed from the data.

The result of the analysis using macro-statistics-based innovation investment data are as follows. The scale of innovation investment is gradually expanding in all industries, manufacturing and service industries. Also, the manufacturing sector shows a relatively high R&D concentration compared to the service sector. Furthermore, looking at the average annual investment growth rate by types of innovation investment, R&D investment and software investment have increased on average in most manufacturing and service industries, whereas organization investment shows a negative average annual investment growth rate in both industries. On the other hand, the trend of the average annual growth rate varies depending on the characteristics of industries. Investment in human resource contracted during the sample period in most industries of manufacturing sector, but the investment polarization pattern among the industries can be seen in service sector. In case of IP investment, the polarization phenomenon was evident due to difference of the nature of the industries within sectors. Also, in the case of marketing investment, the industries of service sector shows an increasing pattern while the industries of manufacturing sector shows a decreasing pattern.

The results of the analysis using micro-statistics-based innovation investment data are as follows. The scale of innovation investment in all domestic industries has increased and its ratio to sales in 2020 is slightly higher than that in 2001. Innovation investment in the manufacturing sector increased to 4 times compared to 2001 and the ratio to sales increased to 3.2%. As the proportion of investment in the top 10% of firms insales has decreased,theproblemofimbalance inthe investment structure is somewhat alleviated.

The results of examining the structure of innovation investment by dividing the manufacturing industry into three types - the ICT, machinery, and materials sector - are follows. The scale of innovation investment expanded the most in the ICT sector, followed by machinery and materials. The ratio of innovation investment to scales was also the largest in the ICT sector, so it was confirmed that the investment propensity was higher in this sector thanthat of other sectors.Inaddition, R&D investment has a structure that is about six times higher than non-R&Dinvestment. This is characterized by animbalanced relationship between R&D and non-R&D investment. The result of examining the structure of innovation investment by dividing the manufacturing industry by technology levels are follows. The scale of innovation investment and the ratio to sales expanded the most in hightechnology group, followed by low technology and middle technology group. In the mid-tech group, R&D investment took up to about 83% of the total investment, showing a relatively large proportion. Also, it can be seen that only a small number of firms makes innovation investment with large scales.

When looking at the service industry by three types, the innovation investment in producer and social/personal service sector has increased by about 2.6 times, and distribution service sector has expends the innovation investment by 4.2 times during sample period. However, innovation investment to sales ratio in all three sectors stays under 1%, which implies that the expenditure on innovation investment by service firms is insignificant. The proportion onR&Dinvestment was the highest in producer service sector, and in non-R&D investment, marketing investment was largest in all three service sectors.

Meanwhile, when analyzing innovation investment by firm size, it shows that the amount ofinnovationinvestment insmall andmedium-sized enterprises(SMEs) increased relatively quickly and concentrated onR&D investment. In the manufacturing sector, the scale of innovation investment expends in all firm size and most investment is heavily concentrated onR&Dinvestment. In service sector, the ratio of innovation investment to sales is in only 0~1% range for all firmsizes, which implies a significantly week level. The ratio of R&D investment is relatively higher than that of non-R&D investment and in non-R&D investment, marketing investment is relatively large.

In addition, we attempts to identify firm’s perceptions of innovation investment, difficulties, and policy-related demands by conducting a survey. First, it is found that there is a gap between the innovation investment perceived by firms and the innovation investment actually implemented. These results could be due to the lack of awareness of the effectiveness and efficiency of the various innovation investment. Also,the internal difficulties facedby firms include the lackof professionals and the burden of financing, and the external difficulties include burden of external financing and the inadequacy of the support system. Furthermore, it was found that the awareness of innovation investment support program was low, and the proportion of firms with experience in using support program was even lower. Therefore, the government should design policies in consideration of the difficulties that firms face in implementing innovation investments.

The discussions on various innovative investment and related policies have continued, and its necessity has been re-emerging with the recent economic structural change. However, the current innovation investment support program is heavily concentrated on R&D investment. Thus, for successful innovation, various innovation investment must be made in parallel withinvestment inR&D.Inthis regard, the structureofinnovation investment focused only on R&D investment can rather impair R&D efficiency. Therefore, in order to increase the success rate of innovation activities, it is necessary to examine the economic contribution of each type of innovation investment and to determine the size and availability of policy support for each type of innovation investment based on the research results.