In this study, I analyze the effects that supply-networks between LF (large firms) and SME (small & medium enterprises) have on SME investments, both TAI (Tangible Asset Investment) and RDI (Research & Development Investment) utilizing Panel Data provided by KED (Korea Enterprise Data). For that purpose, I distinguish the supplier SMEs that are supplying to flagship firms through establishing the transaction relationship between LF and SME from market suppliers and then compare their RDI and TVI.
The LF’s outsourcing strategy can have enormous effects on SME investment. In case where the LF outsources the middle part production of the value-chain which has low added value, the SMEs will be less tempted for RDI, whereas their TAI to supply the intermediary goods may increase. In this case, the supply-network between LF and SME might lead to the SME’s growth. However, the technological gap between LF and SME may widen even more. Also, if the SMEs’ profits generated from their investment belong to LF demanders, the suppliers’ investment will lower in its ratio than that of market suppliers that do not take part in the supply-network between LF and SME. Much existing research suggests that the subcontract deals between LFs and SMEs increase the suppliers’ turnover, but the profitability itself of the suppliers does not increase.
The followings are the results from my analysis. First, the RDI of SMEs that participate in transactions between LF and SME is significantly lower than RDI of market suppliers that are not part of any transaction between LF and SME. With the sole exception of the high-tech industry, in every other industry the supplier SMEs have lower levels of RDI (R & D Intensity) than that of market suppliers outside the network with the LFs.
Secondly, I analyze the effects that the supplier SMEs’ transaction concentration has on their RDI to verify whether the reason the supplier SMEs’ RDI is less than that of market suppliers outside the network is because of their specificity or not. My result from the pooled OLS model demonstrates that the relationship between HHI(Herfindahl-Hirschman index) and RDI is of an inverted U type.
However, the model where I controlled the “within effects” of the panel units does not give me any statistically significant results. Generally, at the low transaction concentration level, since the supplier SMEs’ dependency on their demanders is relatively low, the increase of their supplying has appeared to promote their RDI. Nonetheless, after the transaction concentration exceeds a certain point, the transaction concentration with flagship firms has negative effects on the suppliers’ RDI. Although the same result does not arise in the model of the within effects of the panel units controlled, I cannot interpret the increase of HHI in each supplier as a factor in increasing their RDI.
Thirdly, the estimates of TAI show differences in many aspects from the estimates of RDI. The results from TAI analysis do not show any statistically significant differences between the TAI of the supplier SMEs within the network between LF and SME and of those out of it. This also holds true in analysis on each industry. In all industries, there is no statistically significant difference in TAI-sail Ratio between the suppliers within the network between LF and SME and those outside it. This means that the transaction between LF and SME does not force the suppliers to carry out more investment, but at the same time, it does not reduce the burden of investing as well.
Fourth and last of all, from the estimates of transaction concentrating only of the supplier SMEs, independently from analysis models used, the estimates of the transaction concentrations turn out to be positively definite. However, as shown in the RDI analysis, the relationship between the HHI and TVI is not an inverted U type. This means that the higher the transaction concentration is, the higher the supplier SMEs’ TVI is.
From the analysis on the hierarchical supplier stages, I discovered that only at the first stage do supplier SMEs increase their TVI as their transaction concentration increases. Therefore, it is safe to say that out of all the supplier SMEs it is only the first stage supplier SMEs that are affected by the transaction concentration. It is assumed that the other supplier SMEs expand their production through rising their facility utilization rate when their transaction concentration increases. Moreover, out of all the industries, only the medium-high-tech industry is the one that shows a significant effect of transaction concentration on TVI.