The recent academic literature has demonstrated empirical trends pointing towards declining business dynamism in the US (Autor et al. 2017, Akcigit and Ates 2019a,b). Market concentration, firms’ profit shares and mark-ups have been rising while the labour share, job reallocation rates and firm entry have been declining. Put together, these empirical facts can give a remarkably dense overview of the state of an economy and business dynamism. Akcigit and Ates (2019a,b) show how these diverse facts pointing to declining business dynamism in the US can be consistently explained within an endogenous growth model framework. Though the reasons behind declining business dynamism are likely to differ across countries, similar findings have been established for a number of OECD countries (Andrews et al. 2015). Academic and policy discussions continue about how widespread the trend in declining business dynamism is and what the causes in each country are.
Facts on business dynamisms in Turkey
In a recent paper, we establish a number of facts pertaining to business dynamism in the Turkish manufacturing sector for the decade between 2006 and 2016 (Akcigit et al. 2019c).1 We find that business dynamism was stable or even improving until 2012 but began to deteriorate afterwards. Figures 1 and 2 illustrate the decline in business dynamism after 2012. Figure 1 shows that market concentration, defined as the average industry employment share of the largest 4 or 20 firms, declined until 2012 but entered a rising trend since then. Figure 2 presents a dramatic decline in the share of employment of young firms in the post-2012 period. The decline in business dynamism illustrated by these two figures coincides with the decline in average employment growth of Turkish firms, shown in Figure 3. The decline in employment growth that we find in the firm data is confirmed by the rise in the unemployment rate since 2012.
Figure 1 Market concentration between 2006 and 2016
Figure 2 Share of employment in young firms
Figure 3 Average firm employment growth
We further establish a number of empirical facts on mark-ups and profits, job reallocation, labour share, productivity growth and frontier firms. Practically all empirical findings point to a break since 2012:
- Fact 1. Firm average employment growth has declined, and unemployment has risen
- Fact 2. Market concentration has increased
- Fact 3. Mark-ups and profits have increased
- Fact 4. Average productivity of frontier (i.e. highly productive) firms has not increased
- Fact 5. There is increasing persistence among incumbent frontier firms
- Fact 6. Firm growth rate dispersion has decreased
- Fact 7. Job reallocation has declined
- Fact 8. Increase in market shares is associated with lower labour shares
- Fact 9. Firm entry rate has declined and exit rates have increased
- Fact 10. Employment shares of young firms has declined
Of special interest is Fact 4, which states that the average productivity of frontier firms has not increased unlike the US. Figure 4 shows this fact. The black line shows the average productivity of the top 5% of the productivity distribution while the dashed line shows the average productivity of the other 95%. Given that market concentration is rising, it is natural to assume that this is due to a growth in the productivity gap between top firms and their followers. However, Figure 4 shows that this is not the case. There is no increase in the productivity gap between frontier and laggard firms in Turkey. The decline in business dynamism does not seem to be the product of a leap in productivity by market leaders.
Figure 4 Frontier and laggard firm productivity
An endogenous growth model
We introduce an endogenous growth model in the paper to understand the mechanisms behind the empirical trends we observe in Turkey. The trends we find are consistent with a negative shock to the research and development (R&D) investment costs of follower firms that can credibly challenge sector leaders. If follower firms are unable to credibly challenge sector leaders, there should be a rise in market concentration and mark-ups without an increase in the productivity of market leaders. The intuition is simple. If follower firms are unable to innovate due to higher investment costs, there will be no competitive pressure on sector leaders to improve their productivity.
The obvious policy suggestion is to compensate for the negative shock to the R&D investments of follower firms by subsidising their R&D investments, which will eventually put pressure on market leaders to innovate. However, the key question that remains is what the negative shock to R&D investments might be. In the next section, we briefly discuss a potential negative shock to the financing of investments of follower firms.
Tighter credit markets and business dynamism in Turkey
Like many of its developing country counterparts, the Turkish economy relies heavily on foreign capital for investment and economic growth. Dependence on foreign capital is particularly acute in Turkey due to structurally low saving levels. The health of the financial sector is therefore a main determinant of the state of the overall macroeconomic outlook. After a severe financial crisis in 2001, Turkey adopted a wide range of financial and fiscal structural reforms as part of its economic stabilisation programme. The 2009 financial crisis had immediate negative effects. However, the reform programme after 2001 ensured that the economy bounced back in 2010 and 2011 with high GDP growth rates, as it received a large share of the low interest rate credit made available to developing economies from the monetary easing in developed countries. However, beginning with the Fed tapering in May 2013, Turkish credit markets began to tighten. Figure 5 shows that the decline in credit availability was not equal across Turkish firms, as the available credit also became more concentrated among fewer firms since 2013.
Figure 5 Credit concentration
Figure 6 Credit concentration by source
To understand whether the rising concentration in credit markets is caused by a change in foreign credit flows, Figure 6 shows the degree of concentration in foreign currency denominated credit (FX credits) along with a further breakdown of FX credit into that issued by domestic banks and by external foreign financial institutions (cross-border FX credits). The rise in the concentration of cross-border FX credits suggests that only market leaders are able to obtain credit in the foreign credit markets after 2012. This may well be the negative shock to investment costs of follower firms that was discussed in our theoretical model. If follower firms face higher interest rates or are simply unable to find enough credit post 2012, their ability to put pressure on market leaders will be limited and consequently, there will be a decline in business dynamism. Nevertheless, the finance channel is only one of several plausible explanations for the decline in business dynamism in Turkey, and further research is certainly needed to explore alternative channels such as uncertainty.
The microeconomic dynamics show that the macroeconomic situation in Turkey has changed since the beginning of 2012, and this trend is led by a decline in business dynamism. The root cause is likely the inability of follower firms to credibly challenge market leaders in Turkey. As a potential explanation for the break in the trend observed in 2012, we offer the finance channel. Its shows an increase in the concentration of FX credit issued to Turkish firms, particularly the cross-border FX denominated credit. Regardless of the cause, our theoretical model gives a consistent explanation to all empirical facts we established, which suggests a path forward through subsidies towards the R&D investment of medium to large follower firms that can credibly challenge and put pressure on the market leaders.
Akcigit, U and S T Ates (2019a), "Ten facts on declining business dynamism and lessons from endogenous growth theory”, American Economic Journal: Macroeconomics, forthcoming.
Akcigit, U and S T Ates (2019b), “What happened to U.S. business dynamism?”, NBER working paper 25756.
Akcigit, U, Y E Akgunduz, S M Cilasun, E Ozcan-Tok and F Yilmaz (2019c), “Facts on business dynamism in Turkey”, Central Bank of the Republic of Turkey Working Paper 19/30.
Andrews, D, C Criscuolo and P N Gal (2015), “Frontier firms, technology diffusion and public policy”, OECD Productivity Working Paper No. 2.
Autor, D, D Dorn, L F Katz, C Patterson and J Van Reenen (2017), “Concentrating on the fall of the labor share”, American Economic Review 107(5): 180–85.
 We find qualitatively similar trends in the full economy for most business outcomes. Full economy trends can be found in the appendix section of Akcigit et al. (2019c).