This Selected Issues paper on Turkey assesses the role of structural reforms in enhancing productivity growth in advanced and emerging economies and discusses results that are relevant for Turkey. The paper investigates the role of structural reforms in boosting productivity growth and describes the stochastic frontier set-up for analyzing factors that affect output through technical efficiency; and subsequently presents empirical results. It also simulates productivity gains from closing the structural reform gaps between Turkey and its benchmark. Structural reforms to improve hiring and firing regulations, the business and regulatory environment, and skills are found to have the largest estimated long-term productivity gains for Turkey. In order to bolster Turkey’s sustainable medium-term growth prospects, structural reforms should be implemented sooner rather than later, and any possible negative reform impacts in the short run could be limited by a reform sequencing and reform complementarities.

Abstract

This Selected Issues paper on Turkey assesses the role of structural reforms in enhancing productivity growth in advanced and emerging economies and discusses results that are relevant for Turkey. The paper investigates the role of structural reforms in boosting productivity growth and describes the stochastic frontier set-up for analyzing factors that affect output through technical efficiency; and subsequently presents empirical results. It also simulates productivity gains from closing the structural reform gaps between Turkey and its benchmark. Structural reforms to improve hiring and firing regulations, the business and regulatory environment, and skills are found to have the largest estimated long-term productivity gains for Turkey. In order to bolster Turkey’s sustainable medium-term growth prospects, structural reforms should be implemented sooner rather than later, and any possible negative reform impacts in the short run could be limited by a reform sequencing and reform complementarities.

Productivity Payoffs of Structural Reforms in Turkey

A. Introduction

1. Turkey’s economy has grown rapidly over the past decade, mainly by increasing inputs. With an average annual real growth of around 5½ percent over the last 10 years, Turkey has been one of the fastest growing economies in G20 and other larger emerging markets. However, growth has become more unbalanced. Strong post-2009 output growth has been largely due to increased capital accumulation, driven by construction investment activities and financed, in large part externally, and by rapid credit growth. Labor inputs also contributed positively to growth, supported by a pick-up in female labor force participation—but its contribution appears to have leveled off in recent years. Over the same period, the high investment rate (greater than the neoclassical golden rule benchmark would suggest), together with high inflation and elevated private sector and external debt, point to past over-investment.1

uA01fig01

Turkey: Factor Contributions to Potential Output Growth (Percentage points)

Citation: IMF Staff Country Reports 2019, 396; 10.5089/9781513524665.002.A001

Sources: TurkStat; Presidency of Strategy and Budget; and IMF staff calculations.

2. Growth prospects have weakened due, among other things, to declining total factor productivity (TFP) growth, calling for well-prioritized productivity-enhancing structural reforms. The high contributions of factor inputs is consistent with average annual TFP growth declining from 1.9 percent in 2002-06 to around zero in 2010-18. As the potential of input-led growth is slowing, structural reforms are needed to boost productivity and Turkey’s medium-term growth potential. Nonetheless, as in many other countries, undertaking structural reforms can be economically and politically challenging including due to their possible temporary costs.2 Prioritizing reform measures and packaging reforms to benefit from their complementarities are therefore key to the success and durability of reforms.

3. This paper will assess the role of structural reforms in enhancing productivity growth in advanced and emerging economies, and discuss results that are relevant for Turkey. The paper is structured as follows. To investigate the role of structural reforms in boosting productivity growth, Section B first describes the stochastic frontier set-up for analyzing factors that affect output through technical efficiency; and subsequently presents empirical results. To highlight policy priorities for Turkey, the section also simulates productivity gains from closing the structural reform gaps between Turkey and its benchmark. Section C discusses policy implications and concludes.

B. Technical Efficiency Gains from Structural Reforms

Empirical Framework

4. To estimate the impact of structural reforms on productivity growth, TFP growth is decomposed into contributions from common technological change and a country-specific technical efficiency; and is estimated using a stochastic frontier analysis (SFA) approach. Following Cardarelli and Lusinyan (2015); and Lusinyan (2018), the level of output for country i at time t, Yi,t, can be written as:

Yi,t={f(Xi,t;β)exp(νi,t)}θi,t(Zi,t;δ)(1)

where {f(Xi,t; β) · exp(νi,t)} is the country-specific efficiency frontier - in which Xi,t denotes the quantities of inputs (capital and labor); β denotes the vector of parameters that define the production function, f(·); and νi,t is iid N(0,σν2) random shocks. θi,t(Zi,t; δ) is the country-specific distance of the actual output from the efficiency frontier (or the technical efficiency), ranging between (0,1], whereby θi,t = 1 indicates that the country produces the optimal output at the efficiency frontier.3 The technical efficiency can in turn be described as a function of structural variables, Zi,t, with corresponding parameters, δ.

As opposed to the traditional way of regressing TFP growth, proxied by a Solow-type residual, on structural reforms variables, the SFA approach allows for a simultaneous estimation of the parameters of the stochastic frontier production function, and the model for technical efficiency using a maximum likelihood method (Battese and Coelli, 1995; and Belotti et al, 2013). Using a loglinear Cobb-Douglas production function with capital (Ki,t) and labor (Li,t) inputs, Equation (1) can be rewritten as:

Efficiency Frontier:ln Yi,t=β0+βK ln Ki,t+βL ln Li,t+νi,tui,t(2)
Model of Inefficiency:ui,t=δ0+δZ ln Zi,t+wi,t(3)

where ui,t = —ln θi,t is the country-specific inefficiency. The point estimate of technical efficiency can then be calculated as E(exp(— ui,t|ei,t) with ei,t being the model’s error term comprised of the two independent, unobservable, error terms, νi,tui,t.

5. The efficiency frontier and technical inefficiency are estimated over cross-country panel data, using a broad range of structural reform variables.4 A sample of about 110 advanced and emerging economies covering 1990-2017 is constructed for the analysis. Macroeconomic variables—such as real output, stock of capital, employment, and output gap—are obtained from the IMF’s World Economic Outlook (WEO) database, the Penn World Table, and the World Bank’s World Development Indicators (WDI). A number of data sources are used for structural reform variables covering the areas highlighted by the literature—such as business regulations, labor market (rigidity, wage-setting, skills), and domestic and foreign competition in the product markets.5 The structural reform variables used in the regressions are those available for a relatively longer time span (See Appendix I for data description and sources). Correlations between structural variables from different sources are also analyzed and presented as robustness checks.

Empirical Results

6. Indicators of the business and regulatory environment, as well as labor and product markets are found to affect technical efficiency. The results of the SFA regressions using the aggregate indicators of structural reforms are presented in Table 1, with the first sub-panel showing the estimated frontier production function and the second sub-panel showing the simultaneously estimated model of inefficiency. Better regulatory quality to support private sector development and higher competition in the product markets are associated with higher efficiency (negative impacts on inefficiency). Labor market flexibility, as well as education and training, also help reduce inefficiency. In terms of the magnitude of the impact, labor market flexibility seems to be associated with the largest efficiency gains, followed by competition in the product markets, regulatory quality, and education.

Table 1.

Stochastic Frontier Analysis with Conditional Inefficiency— Broad Structural Reform Indicators 1/2/

article image

* p<0.1 ** p<0.05 *** p<0.01.

Clustered standard errors (at the income-group level) in parentheses.

7. Using sub-indicators of these broad reform categories provides a robustness check and highlights specific aspects of labor and product markets that are key to enhancing efficiency. Table 2 displays the results when inefficiency is estimated as a function of detailed structural indicators, which are sub-components of labor market flexibility, product market competition, and education and training.

Table 2.

Stochastic Frontier Analysis with Conditional Inefficiency— Sub-indicators of Structural Reforms 1/2/

article image

* p<0.1 ** p<0.05 *** p<0.01.

Clustered standard errors (at the income-group level) in parentheses.

PISA stands for Program for International Student Assessment.

  • Among various aspects of labor market flexibility, the analysis suggests a relatively higher efficiency impact from rigidity related to hiring and firing practices, followed by alignment between compensation and productivity. Although good worker-employer relations and flexibility of the wage-setting mechanism are found to have a positive effect on efficiency, the effects are small and/or statistically insignificant.

  • In terms of regulatory environment and competition, variables related to domestic competition— such as the extent of market dominance and time required to start a business—play a larger role in enhancing efficiency than those related to foreign competition—such as trade barriers.

  • Related to human capital, quality of education and educational outcomes are positively associated with higher efficiency. However, the effect of on-the-job training in reducing inefficiency is not statistically significant.

8. Structural reform priorities for Turkey can be identified by the expected gains from closing existing structural policy gaps. Turkey’s structural policy gaps are measured by comparing the country to the 75th percentile of OECD countries, and the impacts on long-run real GDP growth are simulated for a scenario that assumes that Turkey moves closer to the OECD benchmark. Specifically, as most reforms take time and their effects typically materialize gradually (see for example Dabla-Norris et al, 2015; Bordon et al, 2016; IMF, 2016; and IMF, Forthcoming), the simulation assumes that the structural policy indicators for Turkey converge towards the chosen OECD benchmark over a twenty-year period. Figure 1 presents both the policy gaps between Turkey and the OECD benchmark and the expected long-run gains from closing half of the gaps based on the SFA regression results from Tables 1 and 2.

Figure 1.
Figure 1.

Structural Reform Gaps and Payoffs Through Higher Efficiency

Citation: IMF Staff Country Reports 2019, 396; 10.5089/9781513524665.002.A001

Sources: IMF staff estimates.Note: For each policy variable, the scatter plot presents Turkey’s distance from the 75th percentile of OECD countries, and the estimated equivalent increase in annualized output levels if half of the gap is closed in the long run. Figure 1-a is consistent with econometric results from Table 1, and Figure 1-b is consistent with econometric results from Table 2.
  • Turkey lags further behind the OECD benchmark in the areas of regulatory quality, quality of education, and worker-employer relations (Figure 1-b, on the x-axis). Meanwhile, the structural policy gaps related to hiring and firing regulations and the extent of market dominance are moderate, and there’s only a small gap on foreign competition, measured by the prevalence of trade barriers.

  • The simulations show that reducing the gaps in regulatory quality and adopting more flexible hiring and firing regulations are likely to have the strongest impact on technical efficiency (Figure 1-b), raising annualized output levels by around 0.3 percentage points over a 20-year period. Moving towards the OECD benchmark on quality of education would be associated with an additional annualized output levels of 0.2 percentage points. Meanwhile, reducing the extent of market dominance and improving pay-productivity alignment and cooperation in worker-employer relations towards the OECD benchmark would also yield some, albeit smaller, output gains.

9. The empirical results should not be viewed as precise payoffs, but rather be used to guide reform prioritization. Due to potential endogeneity problems and also relatively smaller gains found in the recent literature (IMF, 2019), these empirical results should be interpreted with caution. Nevertheless, the relative payoff magnitudes across different reforms could help inform policy prioritization.

Cross-checking Structural Reform Variables from Different Data Sources

10. To minimize bias or other methodological limitations attached to specific structural reform indicators alternative indicators from other sources are also used. While the regression analysis relies mainly on structural reform indicators from the World Bank and the World Economic Forum due to their availability, the correlations between these indicators and similar indicators from other sources—such as the OECD, the Fraser Institute, and the World Bank’s Doing Business—are presented in Appendix II. Most structural reform indicators from various sources are strongly correlated with each other, especially those related to business and regulatory environment and labor market conditions, suggesting that the results are generally robust to variable choice and source.6

C. Policy Implications and Conclusion

11. Structural reforms to improve hiring and firing regulations, the business and regulatory environment, and skills are found to have the largest estimated long-term productivity gains for Turkey. Closing each of these policy gaps by half, relative to the OECD benchmark, is associated with output gains equivalent to an increase in annual real GDP growth of around 0.2-0.3 percentage points. Meanwhile, other structural reforms related to both labor and product markets—such as enhancing competitiveness in the product market, better aligning pay and productivity, and improving worker-employer relations—are also expected to yield positive gains equivalent to about 0.1 percentage point increase in the real GDP growth rate.

12. To bolster Turkey’s sustainable medium-term growth prospects, structural reforms should be implemented sooner rather than later, and any possible negative reform impacts in the short run could be limited by a reform sequencing and reform complementarities. Although the dynamic impact of structural reforms and reform complementarities are beyond the scope of this paper, some other studies find that the short-run negative impacts can be mitigated as follows.7

  • Given short-term macroeconomic challenges, product market reforms could deliver short-term gains that do not depend strongly on the economic cycle, and hence should be undertaken early. Product market efficiency could be enhanced by simplifying business entry and exit and addressing administrative and regulatory barriers to competition (OECD, 2016). In addition, energy prices could follow an automatic pricing mechanism, which would help improve efficiency and contain contingent sovereign liabilities. On the other hand, reforms related to labor markets should be carefully calibrated, especially on job protection, due to potential short-run negative impacts when economic conditions are not as strong.8

  • In addition, the reform-growth relationship is highly heterogeneous and could be influenced by the country’s institutional environment (Prati et al, 2013). Although it was relatively more difficult to advance, improvements in governance seem to have been associated with higher payoffs of structural reforms in some Eastern European countries during the 1990s through mid-2000s (Roaf et al, 2014). IMF (2019) also finds that the quality of governance matters for the magnitude of gains from structural reforms. Particularly in emerging markets and low-income countries, structural reforms can deliver large gains where governance is strong; but will be less successful in paying off where governance is weak.

Appendix I. Data Description and Sources

article image
article image

Appendix II. Cross-checking Structural Reform Variables

uA01fig02

Business and Regulatory Environment Variables

Citation: IMF Staff Country Reports 2019, 396; 10.5089/9781513524665.002.A001

Sources: WB’s World Governance Indicators; and Economic Freedom of the World (Fraser Institute).Sources: WB’s World Governance Indicators; and World Economic Forum.Sources: WB’s World Governance Indicators; and World Economic Forum.Sources: WB’s World Governance Indicators; and Doing Business.
uA01fig03

Government Efficiency Variables

Citation: IMF Staff Country Reports 2019, 396; 10.5089/9781513524665.002.A001

Sources: WB’s World Governance Indicators; and World Economic Forum.
uA01fig04

Labor Market Flexibility Variables

Citation: IMF Staff Country Reports 2019, 396; 10.5089/9781513524665.002.A001

Sources: World Economic Forum; and Economic Freedom of the World (Fraser Institute).Sources: World Economic Forum; and Economic Freedom of the World (Fraser Institute).Sources: World Economic Forum; and Economic Freedom of the World (Fraser Institute).Sources: World Economic Forum; and OECD.
uA01fig05

Education Variables

Citation: IMF Staff Country Reports 2019, 396; 10.5089/9781513524665.002.A001

Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.
uA01fig06
uA01fig06

Product Market Regulation and Competition Variables

Citation: IMF Staff Country Reports 2019, 396; 10.5089/9781513524665.002.A001

Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.Sources: World Economic Forum; and OECD.

References

  • Banerji, A., V. Crispolti, E. Dabla-Norris, R. Duval, C. Ebeke, D. Furceri, Tl. Komatsuzaki, and T. Poghosyan, 2017. “Labor and Product Market Reforms in Advanced Economies: Fiscal Costs, Gains, and Support,” IMF Staff Discussion Note 17/03, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Belotti F., S. Daidone, G. Ilardi, and V. Atella, 2013. “Stochastic Frontier Analysis Using Stata,” Stata Journal, 13 (4), pp.719-758.

    • Search Google Scholar
    • Export Citation
  • Biljanovska, N., and D. Sandri, 2018. “Structural Reform Priorities for Brazil,” IMF Working Paper 18/224, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Bordon, A. R., C. Ebeke, and K. Shirono, 2016. “When Do Structural Reforms Work? On the Role of the Business Cycle and Macroeconomic Policies,” IMF Working Paper 16/62, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Bouis, R., O. Causa, L. Demmou, and A. Zdzienicka, 2012. “The Short-Term Effects of Structural Reforms: An Empirical Analysis,” OECD Economics Department Working Papers, No. 949, OECD Publishing, Paris.

    • Search Google Scholar
    • Export Citation
  • Bluedorn, J. C., S. Aiyar, R. A. Duval, D. Furceri, D. Garcia-Macia, Y. Ji, D. Malacrino, H. Qu, J. Siminitz, and A. Zdzienicka, 2019. “Strengthening the Euro Area; The Role of National Structural Reforms in Building Resilience,” IMF Staff Discussion Notes 19/05, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Cacciatore, M., R. Duval, and G. Fiori, 2012. “Short-Term Gain or Pain? A DSGE Model-Based Analysis of the Short-Term Effects of Structural Reforms in Labour and Product Markets,” OECD Economics Department Working Papers, No. 948, OECD Publishing, Paris.

    • Search Google Scholar
    • Export Citation
  • Caldera-Sánchez, A., A. de Serres, and N. Yashiro, 2016. “Reforming in a difficult macroeconomic context: A review of the issues and recent literature,” OECD Economics Department Working Papers, No. 1297, OECD Publishing, Paris.

    • Search Google Scholar
    • Export Citation
  • Cardarelli R. and L. Lusinyan, 2015. “U.S. Total Factor Productivity Slowdown; Evidence from the U.S. States,” IMF Working Paper 15/116, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Dabla-Norris, E., G. Ho, and A. Kyobe, 2016. “Structural Reforms and Productivity Growth in Emerging Market and Developing Economies,” IMF Working Paper 16/15, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Romain D., and D. Furceri, 2018. “The Effects of Labor and Product Market Reforms: The Role of Macroeconomic Conditions and Policies,” IMF Economic Review, Col. 66(1), pp.31-69, March.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2015. “Structural Reforms and Macroeconomic Performance: Initial Considerations for the Fund,” IMF Policy Paper, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2016. “Time for A Supply-side Boost? Macroeconomic Effects of Labor and Product Market Reforms in Advanced Economies,” World Economic Outlook, April 2016: Chapter 3, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2018. “Turkey: 2018 Article IV Consultation—Staff Report,” IMF Country Report 18/110, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2019, “Reigniting Growth in Low-income and Emerging Market Economies: What Role Can Structural Reforms Play,” World Economic Outlook, October 2019: Chapter 3, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Lusinyan L., 2018. “Assessing the Impact of Structural Reforms Through a Supply-side Framework: The Case of Argentina,” IMF Working Paper 18/183, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • OECD, 2016. “Turkey Policy Brief,” OECD.

  • OECD, 2018. “OECD Economic Surveys—Turkey”, OECD.

  • Prati, A., M. G. Onorato, and C. Papageorgiou, 2013. “Which Reforms Work under What Institutional Environment? Evidence from a New Data Set on Structural Reforms,” Review of Economics and Statistics, 95 (March), pp. 946-968.

    • Search Google Scholar
    • Export Citation
  • Roaf, J., R. Atoyan, B. Joshi, K. Krogulski, and an IMF Staff Team, 2014. “25 Years of Transition: Post-Communist Europe and the IMF,” Regional Economic Issues Special Report, October 2014, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • World Bank, 2019. “Country Economic Memorandum: Firm Productivity and Economic Growth in Turkey,” World Bank, Washington DC.

1

IMF Country Report No. 18/110, Annex I.

2

See for example Bouis et al (2012), Cacciatore et al (2012), Banerji et al (2016), and IMF (2016).

3

To focus on the role of country-specific structural reforms, the common technological change is assumed to be zero, as is typically assumed in the existing literature.

4

In all regressions, the output gap and output volatility variables are also included in the model of inefficiency to control for cyclical variation which could impact efficiency.

5

See for example, Prati et al (2013), IMF (2015, 2016), Bordon et al (2016), Biljanovska and Sandri (2018), and Lusinyan (2018). Some of these studies also include variables capturing structural reforms related to the financial sector, such as restriction to bank competition and access to finance. In most SFA specification, the coefficient estimates of these variables are statistically insignificant (results not shown here), and hence are dropped from the analysis.

6

The correlations between some product market reform indicators from different sources are slightly weaker, potentially due to a smaller number of observations (as indicators from alternative sources span over a much shorter time period). However, they remain statistically significant.

7

See, for example, Bouis et al (2012), IMF WEO Chapter (2016), Duval and Furceri (2018), and IMF WEO Chapter (Forthcoming) for discussions on the dynamic impact of structural reforms.

8

It is important to note that, in some cases, even with weak demand and limited fiscal and monetary policy support, the short-run negative effects could be reduced by implementing both labor and product market reforms in tandem (Caldera-Sanchez et al, 2016). Product market reforms to reduce entry barriers and enhance stronger competition could help lower prices, increase output and employment, and so reduce the negative impacts of labor market reforms on real wages and employment. A well-communicated, comprehensive, and credible reform package could also help improve business and consumers’ confidence, which could in turn boost consumption today and reduce the need for excessive precautionary savings.

Turkey: Selected Issues
Author: International Monetary Fund. European Dept.