The low productivity of Japan's SMEs has attracted much attention in recent years. David Atkinson, a former partner of US investment bank Goldman Sachs and a member of the government’s growth strategy panel, who is also known to be a key adviser to the Prime Minister, stresses that low productivity of small and medium-sized enterprises (SMEs) is the leading cause of Japan's low growth.1 This discussion is based on the fact that while large businesses have seen a gradual increase in labour productivity since the financial crisis, the productivity of SMEs remains stagnant in Japan (Figure 1) (The Small and Medium Enterprise Agency 2020).
Figure 1 Labour productivity trends in Japan (ten thousand yen)
Source: Financial Statements Statistics of Corporations by Industry (MOF, Japan) data
The government has recognised this situation, and the Ministry of Economy, Trade, and Industry (METI) launched a subsidy programme, the Business Sustainable Subsidy (BSS), in 2013 to resolve the productivity issues of small enterprises. One of its objectives was to increase productivity and to promote more sustainable development of small enterprises through partially supporting business expenses. In our recent paper (Hashimoto and Takahashi 2021), we investigate the effects of the BSS on SMEs’ performance.
Some empirical research does show the more positive effect of subsidies on SMEs than large enterprises. Bronzini and Iachini (2014) evaluate the programme of R&D subsidisation conducted in Italy and show that the subsidies had a positive effect on investment for SMEs, whereas they did not work for large enterprises. Criscuolo et al. (2019) analyse the Regional Selective Assistance (RSA) programme in the UK and find significant positive effects on employment in manufacturing small firms. Policymakers are also interested in these findings on the innovation and job creation potential of SMEs.
In order to evaluate the comprehensive effects of the BSS programme, we test two types of treatment groups: reception and application. While the literature on the effect of receiving subsidies is well established, little attention has been paid to the effect of applying for subsidies on firm performance. This analysis was made possible by using the Tokyo Shoko Research (TSR) data covering 1.5 million firms in Japan and the list of all firms that applied for the subsidies in 2013.
First, we looked at the effect of adoption of the subsidies using the regression discontinuity design (RDD) method, using only the sample of firms that applied for the subsidies. The treatment group consisted of firms that received the subsidy, while the control group consisted of firms that applied for the subsidy but did not receive the subsidy. We could not confirm significant positive or negative effects for firm productivity 1–3 years after the base year. The RDD results reveal that receipt of the subsidy had not led to an improvement in firm productivity.
Next, using a difference-in-difference (DID) model, we estimated the case wherein all applied small firms are the treatment group and not-applied firms comprise the control group. This allows us to examine the effect of the application of the programme itself. In contrast to the reception cases in which we found no significant effects, the results show significant positive effects on the productivity of applicant firms (Figure 2).
Figure 2 The effects of applying for the BSS
Note: This figure shows the estimation coefficients of difference-in-differences model. ***,**,* represents statistical significance at 1%,5% and 10% respectively.
While applying for the subsidy programme has positive effects on performance, it might be that firms with higher performance tend to apply more, suggesting heterogeneity between applicants and non-applicants. The endogenous application decision means that firms' key characteristics, such as business discipline, affect their productivity after receiving the subsidy. To respond to the selection problem for strict causality, we conduct PSM–DID, following Ryan et al. (2019) and choose a pair of samples with similar intervention possibilities. The result shows that even after considering selection bias, applying to the BSS benefits firm productivity.
To examine the heterogeneity of the subsidy application effects by industry, we divided the sample into three industries: construction, manufacturing, and services. For the construction and services industries, the treatment is positively correlated with sales and productivity (Figure 3). Positive sales growth and negative employee growth are also consistent in the services industry findings. We finally show that the services sector contributes a large part of the productivity improvement effect of the BSS application with robustness confirmed through PSM DID.
Figure 3 The effects of applying for the BSS by industry
Note: This figure shows the estimation coefficients of difference-in-differences model by industry. ***,**,* represents statistical significance at 1%,5% and 10% respectively.
We think that these results imply two things. First, subsidies that are small in amount and thinly distributed are less likely to be effective. The subsidy amount in this case is ¥500,000 (€3,800) at most. This amount may be sufficient for small equipment procurement and less expensive promotions; however, it may not be enough to invest in equipment or develop human resources that would lead to fundamental transformation which would increase firm productivity.
Second, the effectiveness of the subsidy application may have been a result of the advisory support embedded in the subsidy scheme. In order to apply for the BSS, firms are obliged to submit their one-year business plan for sales and productivity improvements. In this application process, applicants receive advisory support from local institutions collaborating with the programme to augment their plans. More specifically, firms apply to the Japan Chamber of Commerce and Industry (JCCI) or the Central Federation of Societies of Commerce and Industry (CFSCIJ) for subsidies, depending on the region where they are located. JCCI and CFSCIJ provide consultation services on business management to the applicants. This is a unique feature of the BSS program.
JCCI and CFSCIJ have a total of more than 2,000 chapters nationwide and are familiar with the business environment and issues facing each region. In many cases, they have had long relationships with local member firms. We believe that they are capable of proposing effective business plans for firms applying for the BSS, to increase their productivity. SMEs often demand support from third parties because they lack efficient internal resources. Advisory support helps firms overcome business problems and promotes further growth through the imparting of valuable knowledge (Bennett and Robson 2000). In particular, the effect of external advice may be more pronounced in the services industry, which is relatively less dependent on equipment and more likely to rapidly incorporate outside advice into its management.
Our findings imply that application in itself promotes firms' voluntary activities to address their business issues through external support and improves their productivity. We would like to suggest that it would be helpful for small-scale subsidy policies to incorporate an application process that encourages firms to receive external support. It would improve productivity by allowing firms to address their problems through external advice rather than through the direct provision of financial assistance.
Bennett, R J and P J A Robson (2000), “SME Growth: The Relationship with Business Advice and External Collaboration”, Small Business Economics 15: 193-208.
Bronzini, R and E Iachini (2014), “Are Incentives for R&D Effective? Evidence from a Regression Discontinuity Approach”, American Economic Journal: Economic Policy 6(4): 100‐134.
Criscuolo, C, R Martin, H G Overman and J Van Reenen (2019), “Some Causal Effects of an Industrial Policy”, American Economic Review 109(1): 48-85.
Hashimoto, Y and K Takahashi (2021), “Are Applying for and Receiving Subsidy Worth for Small Enterprises? Evidence from the Government Support Program in Japan”, RIETI Discussion Paper Series, 21-E-039.
Ryan, A M, E Kontopantelis, A Linden and J F Burgess, Jr. (2019), “Now trending: Coping with non-parallel trends in difference-in-differences analysis”, Statistical Methods in Medical Research 28(12): 3697-3711.
The Small and Medium Enterprise Agency (2020), “2020 White Paper on Small Medium Enterprises in Japan” (in Japanese).
1 See https://www.japantimes.co.jp/news/2021/07/05/business/economy-business/japan-minimum-wage/.