OECD Economic Surveys: Turkey 2021

Report

OECD Economic Surveys: Türkiye

14 January 2021

Available in:

English
français

1. Key policy insights

After only partly recovering from a sharp macroeconomic correction in the summer of 2018, Turkey was hit by the COVID-19 shock in spring 2020, slightly later than other countries in the region. While Turkey managed to contain the number of contagion cases relatively effectively in the initial phase, domestic containment measures and the drop in tourism had a severe effect on the economy. Activity contracted, employment fell from an already depressed level after the 2018 shock, and pressures mounted on well-being and social cohesion. Some population groups have suffered more, including informal workers, women, refugees and the youth.

Following the relaxation of containment measures in June and a strong government stimulus, activity rebounded strongly throughout the Summer. Quarter-on-quarter growth was very strong in the third quarter. However, this was followed by a sharp escalation of infections in the Fall, together with rising numbers of fatalities. Pressures on the health system increased again and new confinement measures were introduced from November. Given the elevated uncertainty about the global and local trajectory of the pandemic, and the rapidly increased debt burdens of households and businesses, the recovery is projected to be more gradual than after the previous macroeconomic shocks.

The dynamism of the Turkish business sector has been an asset during the crisis. It has adapted relatively rapidly to the new circumstances, catered to basic domestic needs, and seized new opportunities from international markets. Still, the path of the economy through the pandemic is strewn with strains. They encompass special macroeconomic challenges which arose from the high dependence of the growth pattern on domestic demand and foreign savings, while investor confidence in price stability and policy predictability could not be consolidated and risk premia and exchange rate volatility remained very high. Policy support during the pandemic should be provided in a transparent, predictable and stable macroeconomic framework and without further worsening the external deficit and inflationary pressures. Additional challenges arise from business sector structures, where many firms have small size, weak balance sheets, a narrow equity base, and limited capacity to weather a protracted slowdown and to resume long-term capital formation once a recovery takes hold.

High productivity firms creating good quality jobs remain indeed a minority in the Turkish economy. The largest parts of the business sector still rely on informal or semi-formal practices in employment, corporate governance, financial transparency and regulatory and tax compliance. These appear to have regained ground after the first phase of the COVID-19 pandemic. Ensuring compliance by all businesses with laws and regulations, which should themselves be modernised, will be crucial for their gaining full access to state-of-the-art capital, know-how and technological resources from domestic and international markets, on the way out of the COVID-19 shock and beyond.

This chapter reviews Turkey’s short- and medium-term general economic policy priorities. The subsequent chapter documents how structural change in the business sector can drive post-pandemic growth and social cohesion. The key messages of the Survey are:

After initial successes against the pandemic and a strong economic rebound, Turkey faces a second wave which is putting pressure on the health system, the recovery, the viability of many businesses, employment and social cohesion. This invites the continuation of a supportive policy stance.
Public finances continue to offer room for government support to households and businesses most in need, but this should be provided under a more transparent and predictable fiscal, quasi-fiscal, monetary and financial policy framework. Shortcomings in this framework hindered market confidence in the early phase of the pandemic, creating tensions in risk premia, capital movements and exchange rates which complicated policy responses to the crisis.
New demands and opportunities have emerged for structural change in the business sector. A reform package would help accelerate the ever more needed formalisation of business activities, re-capitalisation of balance sheets, strengthening of investment capacity and digital upgrading of firms of all sizes and sectors. Conditions are now more supportive for constructive social dialogue and participatory adjustments at firm level.

The COVID-19 shock is having a large economic and social impact

Targeted lockdowns and health policy initiatives were initially effective but there is a second wave

The pandemic hit Turkey in the second half of March and diffused at a fast rate. Yet, the number of cases and deaths remained relatively low by international comparison taking into account the size of the population (Figure 1.1). The number of COVID-19 cases peaked at the end of April. The health system faced the pandemic with a low number of physicians and hospital beds per capita (an average of 1.8 physicians and 2.9 hospital beds per 1000 inhabitants, against OECD averages of 3.4 and 4.7), but was well prepared to public health emergencies, thanks, notably, to a strong intensive care infrastructure (with 43.500 intensive care units for a population of 83.4 million). The authorities put in place targeted lockdowns and curfews dedicated to specific age groups, towns and neighbourhoods. International and domestic passenger traffic was entirely shut down. Sectors closed by administrative decision were narrow in international comparison, and not more than 40% of the population was formally confined - except during temporary curfews over weekends and public holidays. Despite this, many activities, particularly those requiring face-to-face interactions, slowed strongly as individuals chose to minimise their health risks.

Intense testing and tracing activities conducted in line with advice from a Scientific Advisory Board were enforced, although there is still room for convergence with international best practices in this area (Figure 1.1, Panel C). According to the OECD Health Policy Tracker, all public and private hospitals were declared ‘pandemic hospitals’ at the height of the crisis, all non-vital elected surgeries were postponed, more than 30 000 additional health professionals were recruited and two new specialised hospitals were constructed in Istanbul - in addition to the large city hospitals recently put in service in many provinces. The existing network of family doctors monitored daily all cases with symptoms (OECD, 2020c). All tests and treatments were financed by the social security administration. Health professionals showed an abnegation and commitment welcomed in all parts of the population (14% of all contagions concerned health workers by August). Masks were made obligatory in all public spaces, and were made available at low cost from the early stages of the pandemic. Saturation was avoided in intensive care units. As in other countries, medical professionals were met with some difficulties in accessing high quality protection gear in certain regions and hospitals (Turkish Medical Association, 2020a), but the number of cases and fatalities were reined in successfully in international comparison in this first phase. Covid-19 vaccine research activities are continuing in national universities and research centers. After a gradual re-opening of the economy from early June (the so-called “return to normal life” measures, which implied a relaxation of lockdowns, re-opening of public spaces, and easing restrictionis of domestic and international passenger transportation), and due to the population’s laxed attitude towards physical distancing, Turkey experienced a steady rise in new cases in summer months (Figure 1.1). The disease spread from densely-populated urban centres to smaller towns and villages. According to one estimate, the coefficient of contagion R0 fell below the critical threshold of 1 in big metropolitan centres by mid-July, but soared above 1 in less densely-populated regions, which was then followed by an upsurge of the coefficient across the entire country (EpiForecasts, 2020).

Available data confirm the vigour of the second wave. After an intermediary peak in mid-September, which proved to be temporary, symptomatic cases and fatalities soared from November, surpassing their April level by a strong margin. The Ministry of Health stated in October that “confirmed COVID-19 cases” were reported according to a narrower definition than recommended by the World Health Organisation, including cases with symptoms but excluding thoses without symptoms (Reuters, 2020a). The Ministry started to publish the number of all confirmed cases from 25 November (Reuters, 2020b). Fatalities continue to be reported according to local definitions. It is essential to uphold confidence in official information on the spread of the pandemic.

In response, a range of measures including changes in school opening plans, restrictions on public events and public space activities, and targeted confinements were introduced. Policies continue to focus on testing, tracing and tracking activities and stronger enforcement of physical distancing. Regional containment measures are managed by provincial authorities. The government introduced a national curfew on certain time slices of the week-ends starting from mid-November. It also extended the curfews already in force for people above 65 and for youth below 20 (with the exception of young workers) to longer time periods.

The new circumstances have serious implications for education. The opening of schools and universities, initially planned for end-September, was postponed for the large part of courses and classes (underpinning the pick-up in Turkey’s policy stringency indexes, Figure 1.1), amid authorities’ efforts to re-open them as quickly as possible. A decision to close them until the end of the year was nonetheless taken on 17 November and was extended to kindergarten on 1 December. The bulk of primary, secondary and tertiary education activities shifted to on-line learning. The impact of school closures on parents’ capacity to resume work, and, more fundamentally, for the quality of education for both school and university students raises important challenges as in all OECD countries (BBC News, 2020a).

Research by the Ministry of National Education (“Evaluation of Distance Learning Activities during the Pandemic”) documented the experience of students, teachers, school principals and parents with on-line education in 2020. More than 800.000 students were surveyed, 11% responded that they were lacking the necessary terminal equipment, 6.7% home or mobile internet access, and around 15% inadequate internet connection (parents’ replies confirmed these proportions). A smaller survey centred on internet access in the Istanbul region (before the creation of on-line support centers and the distribution of free tablet computers by the Ministry, see below) found that 40% of low-income families in the region had no internet access and 58% had no laptop computer (Istanbul Buyuksehir Belediyesi, 2020).

The Ministry took initiatives to mitigate disparities in education access and quality during the pandemic. This included the distribution of 500,000 tablet computers free of charge to disadvantaged children, together with free mobile internet access. Close to 13.900 digital education network support centres and 162 mobile support centers were created through the country, including in schools in disadvantaged areas where students from low-income families could engage in interactive and personalised computer-based learning (EBA, 2020). Guidelines for infection prevention and containment were published for schools and were enforced with trained inspectors. As in all OECD countries, gaps nevertheless emerged between teaching resources and methods between different types of schools. Public schools relied on dedicated television channels (three channels were activated) and on on-line education platforms and live classes. The Ministry agreed with GSM operators to provide cellular subscribers free access to on-line education platforms. The majority of parents found the technical infrastructure of on-line education effective, even if difficulties in internet access remained in certain areas, while 54% were satisfied with its planning. A number of private schools with strong material resources and class-size conditions could implement their own on-line teaching methods. Differences in on-line teaching practices and intensity emerged also between universities. These disparities, if they persist, risk amplifying the social gaps in the quality of education (OECD, 2019a). High-quality studies evaluating the impact of these different streams of on-line teaching on the academic proficiency of students may be required in the future to develop follow-up policies.

Figure 1.1. An intense second wave after the initial success against the pandemic

Strong policy stimulus triggered a vigorous rebound which faced headwinds

The impact of the pandemic on the economy unfolded later than in other countries but was sharp. Activity contracted strongly in April as people chose to limit their interactions, despite limited official restrictions, and external demand declined. Consumption, production and exports all shrank (Figure 1.2). Labour demand fell but the existing short-time work scheme and the new furlough arrangement for unpaid leaves helped to contain job losses in the formal sector. Output fell by 11% in the second quarter of 2020, before recovering strongly in the third quarter.

Informal workers and the self employed were hit the most, as many make their living from contact-intensive services such as retail trade, street catering and public transportation. The high share of these workers in total employment kept the potential for remote working low (at around 20%, against 30 to 40% in other OECD countries (OECD, 2020e)). These groups are not covered by employment-related social safety nets and received only ad hoc cash support. Aggregate household incomes and confidence took a strong hit, reflecting in a sharp fall of private consumption (by more than 25% in April over the previous month according to credit card expenditures). Tourism was hit particularly hard (Figure 1.2 Panel E). This sector employs 7% of all workers, generates demand for a wide range of upstream products and services, especially in certain regions, and accounts for 14% of total exports. Consequently, local and regional demand fell twice as rapidly in touristic regions as in others (Akcigit and Akgunduz, 2020). It is estimated that an expected two-thirds decline in tourism output may have reduced aggregate GDP by about 4% in 2020 (Mehr News Agency, 2020).

Policymakers reacted with a broad set of measures. In the first phase, the fiscal package was relatively limited. The “Economic Stability Shield Programme” announced on 16 March 2020 included 2.1% of GDP of fiscal commitments, including many temporary tax deferrals. The package contained 21 measures accompanied by broad financial and monetary supports (Box 1.1). Emergency aid to households helped avoid situations of extreme distress, but compensated only part of the losses in living standards. Turkey also introduced a series of trade restriction measures (Box 1.1).

Concessional credits to households and businesses played the central role in efforts to uphold demand. They were extended mainly by public banks, but also by private banks incentivised by government guarantees. This increased the share of quasi-fiscal (“below-the-line”) relative to fiscal (“above-the-line”) expenditures. According to the IMF Fiscal Monitor database on specific COVID-19 measures, this share in Turkey was the highest among all countries monitored (Gaspar and Gopinath, 2020).Three public banks were recapitalised.

Box 1.1. Main COVID-19 support measures

On 18 March 2020, the authorities announced a TRY 100 billion (2.1% of GDP) Economic Stability Shield Programme. Complementary measures were added in the following months, and the total amount of measures reached TRY 503.4 billion (10.6% of GDP) as of mid-November. This Box reviews the main measures. Further details are available on OECD’s COVID-19 policy tracker (OECD, 2020c).

Social transfers

From mid-March, the minimum monthly old-age pension was raised from TRY 1.000 to 1.500 (USD 230 on the basis of exchange-rates at the time of announcement).
Families in need received a one-off cash transfer of TRY 1000 (USD 154) per family. By the end of October, 6.3 million families had received the allowance. This transfer was provided as additional support to households receiving other social aid.
Households in need but not eligible for the standard allowance applied for ad hoc support from a new National Solidarity Fund. Public enterprises and private firms were invited to contribute. Around 0.05% of GDP was re-distributed through this fund by September 2020.
Cash and in-kind support was also offered by municipalities. Several municipalities launched local schemes permitting to cancel public utility (water and natural gas) and grocery debts of insolvent households. Private benefactors anonymously closed their accounts payable. These schemes benefitted millions of families.

Concessional credits to businesses and households

In mid-March, the Banking Regulation and Supervision Agency (BRSA): 1) asked banks to postpone their customers’ principal and interest payments for at least three months upon request; 2) extended the delay for classifying a loan as non-performing from 90 to 180 days (in keeping with international recommendations); 3) introduced forbearance measures for the measurement of banks’ Capital Adequacy Ratio; 4) increased the Loan-to-Value Ratio on mortgage loans.
From mid-March, three main public banks (Ziraat, Halkbank and Vakif) offered all businesses concessional working capital loans (at 36 months maturity, 6 month grace period and a low 7.5% interest rate), conditional on their preserving their current employment level.
Public banks offered tradesmen and craftsmen a concessional credit line (at 36 months maturity and 4.5% interest rate). A "craft-and-trade credit card" was also made available under an individual credit line of TRY 25.000 (USD 4.000 at the time of announcement).
Late March, principal and interest payments on SME Bank’s (Halkbank’s) subsidised credits to tradesmen and craftsmen were postponed for three months. Late April, principal and interest payments on Agricultural Bank’s subsidised credits to agricultural producers were postponed for six months.
Public banks started to offer to low-income households (earning less than TRY 5000 – USD 770 per month) “basic need support credits” of up to TRY 10.000 (USD 1.500), with up to 3 year maturity, at a concessional interest rate of 6%.
On 30 March, the Government Credit Guarantee Fund (KGF) increased its total limit for loan guarantees from TRY 25 billion to TRY 50 billion (USD 7.7 billion). It guaranteed general-purpose loans to individuals for the first time. As a result, total loan leveraging capacity of KGF reached TRY 500 billion (14% of 2019 GDP). The ensuing government guarantee liabilities are however capped at 1.4% of GDP.
In April, BRSA introduced an “Asset Ratio” applicable to most banks (except development banks and small banks) to stimulate their credits. Its formulae incited banks to fund new lending from non-deposit sources and/or invest in government bonds. By the end of May the ratio was revised to stimulate longer term loans. In August and September it was revised again to scale-down its expansionary impact. Its phasing out by 31 December 2020 was announced in November (see Box 1.4 for details on the operation of this ratio) .
Early May, the Sovereign Wealth Fund (TVF) injected TRY 21 billion (USD 2.8bn) of additional capital into three public banks enganged in COVID-19 measures (Ziraat, Halkbank and Vakifbank)
At the end of May, these public banks launched an additional set of concessional loan packages to support purchases of domestically produced cars, white goods and other consumer durables. These loans funded also house purchases and domestic holidays.
In June, BRSA increased the upper limit of instalment numbers for credit card purchases from airlines, travel agencies and hotels from 12 to 18 months to stimulate demand for domestic travel and tourism.
In June, the Central Bank (CBRT) introduced a new programme of "Advance Loans Against Investment Commitment". This finances investments that will “reduce imports, boost exports and support sustainable growth” via the recently re-structured Investment and Development Bank of Turkey. Loans are extended with a maximum maturity of 10 year and with an interest rate 150 basis points below the policy interest rate.
Late July, the SME bank (Halkbank) postponed for three months, all capital and interest reimbursements overdue by trademen and craftsmen.
In mid-October, Turkish Banks Association launched a new credit line for tourism firms and their suppliers (“Tourism Support Package”) to finance the wages, rents, and other fixed costs of these enterprises. TRY 10 billlion is made available, to be distributed by 15 banks undergovernment guarantee.
A similar package was introduced in mid-October for SMEs, to help finance their wages, rents and other fixed costs (“Micro Enterprises Support Package”). An envelope of TRY 10 billion will be distributed by three public banks under government guarantee (Ziraat Bank, Halk Bank and Vakifbank).

Monetary support to activity

In mid-March, Turkey’s Central Bank (CBRT) lowered its main policy rate (the one-week repo rate) from 10.75% to 9.75%. It reduced it further to 8.75% on 22 April and to 8.25% on 21 May.
At the end of March, CBRT offered direct liquidity support through 1) an extension of its limits for open market operations on government securities, 2) an extension of the securities accepted as collateral in transactions with banks, 3) an extension of liquidity facilities for banks “for uninterrupted credit flows to businesses”, and 4) an extension of the volume and maturity of its traditional export credits - 70% of this extension is earmarked for SMEs.
In mid-April, CBRT increased its limit for open market operations on government securities from 5% to 10% of its balance sheet.
Early June, it earmarked one third of its total export credit portfolio for long-term investment credits.

Supports to exporters

Turkish Eximbank extended the repayment terms of its existing rediscount credits for exporters by three to six months.
A new Inventory Financing Package by Turkish Eximbank offered low-interest loans to exporters “whose stocks increased due to low demand and canceled orders”.
The maximum specified maturity limit of rediscount credits for exporters were further extended.

Preserving employment links

Eligibility conditions for the existing Short-Time Working Scheme (which compensates 60% of the earnings lost due to shorter work hours) were eased. The requirement of 600 days of contribution was reduced to 450 days, and the need for a valid employment contract in the last 120 days was reduced to 60 days. To be eligible firms should commit not to reduce their employment level. By early November, 3.6 million beneficiaries were paid 21.8 billion TRY under this scheme (Ministry of Family, Labor and Social Services of Turkey, 2020). The application period of this scheme was subsequently extended to 31 December 2020.
The compensatory working period (the re-balancing period for overtime work) was increased from two to four months.
In mid-April, the Parliament adopted a new law on unpaid leaves (furloughs). A fixed monthly allowance of TRY 1170 (USD 170, the floor of unemployment insurance compensation) is granted to furloughed workers. Employers were given discretion in authorising unpaid leaves, in turn they were prohibited from firing any workers during the period the law was in force (it is in force until 17 January 2021 and The President is authorised to prolong it to until 31 July 2021). The workers affected bear nonetheless an income loss compared to their regular earnings. By end-October, 2.1 million beneficiaries were paid around TRY 5.1 billion under this scheme.
From mid-July, a ‘normalisation incentive’ was offered to firms exiting the short-time working scheme. They were exempt from employer and employee social security contributions for six months. By early November, , work contracts of 2.1 million employees were ‘normalised’ via this arrangement. The implementation period of was subsequently extended to 30 June 2021.

Trade Protection

From mid-April, additional customs duties of 2 to 50% were applied to a range of goods, with the goal of “supporting domestic industries adversely affected by the COVID-19 shock”. Resulting net duty rates do not exceed Turkey’s notified World Trade Organisation bound rates. They were to be cut by October, this was subsequently postponed to end-2020.
The list of the products was widened in steps. Around 5000 product groups in total were included (including game consoles, home textiles, white goods, consumer durables, construction materials, industrial machinery, harvesting machinery, textile machinery, sugar confectionery, cocoa powder, chocolate, biscuits, etc.). Surcharge rates gravitate around 17%.
Imports from countries with which Turkey has free trade agreements (FTAs) were exempted. This concerns notably imports from the EU. About 57% of the products subject to surcharges are imported from the EU and other FTA countries, 43% of them are imported from other countries and are affected (including imports from China, India, Japan, Russia and the United States).

Activity rebounded and gained momentum through summer. Credit card spending recovered its pre-shock level in July. House, car and other consumer durable sales, fostered by loan packages, grew sharply. House sales rose to 125% of their level of a year ago, with house prices up by 25%, and even larger increases in certain regions (CBRT, 2020a). The seasonally adjusted PMI index - the reference indicator of business sentiment in Turkey- jumped from a depressed score of 33.4 in April to 56.9 in July (its highest level since March 2011), before declining to 52.8 in September and rebounding to 53.9 in October. Short-time work applications fell, and job vacancy announcements increased. Despite the fall of energy prices, and a still large output gap, inflation responded to domestic demand, stayed close to 12% until October and soared to 14% in November and 14.60% in December.

Exports, despite the weakness of traditional markets in Europe, improved faster than expected. Merchandise exports rebounded by 34% q-o-q in the third quarter of 2020 and reached an all-times high in December. The depreciation of the Lira and the diversification of markets by manufacturers helped. The rebound of industrial activity in Germany - Turkey’s main international value chain customer - played a positive role. Among the main export items, motor vehicles recovered by mid-summer, with also strong growth for textiles and clothing, chemicals and steel products. Manufacturers were active in pandemic-related markets: exports of masks, protective gears and health equipments (including a respiratory assistance device developed by a joint-venture of domestic firms) increased by a total of 530 % in the first half of the year over the same period of 2019.

The upturn was more subdued in services. The traditionally strong correlation between manufacturing and service confidence indexes was broken (Sameks, 2020). Public space activities, including restaurants, leisure services and public transportation stayed frail. E-commerce partially substituted to traditional retail trade. Nearly 40% of Turks were estimated to be using e-commerce at the height of the crisis in April, while the share of on-line sales in total retail sales had approached only 7% at the end of 2019 (Webbrazzi, 2020a and 2020b). The retail franchising association (BMD) reported that nearly 60% of its members achieved a 100% increase in their e-sales between September 2019 and September 2020, but less than one third of them could match their ‘brick-and-mortar’ sale levels of the year ago by the same date (P.A. Turkey, 2020).

Tourism remained very weak, as international visitor numbers fell by 91% over a year ago in July and by 76% in August. Some recovery in domestic tourism, together with Germany, the United Kingdom and Russia freeing up their tourist flows to certain regions of Turkey in August triggered an uptick. However, the United Kingdom reversed their liberalisation decision in late September after the controversy on the accuracy of case reporting, while reservations from Russia (the largest tourism market in terms of visitor numbers) continued to increase (Figure 1.2 Panel E). A study suggested that tourism sector revenues (value added) could fall from USD 34.5 billion in 2019 to USD 15 billion in 2020, before recovering to around USD 25 billion in 2021 (Ernst & Young, 2020a and 2020b).

Balance-of-payment strains have been significant

Turkey faced pressures on its already strained external accounts (following, notably, the 2018 financial turmoil as discussed in the thematic chapter) during the COVID-19 crisis. The worsening of the trade balance in the first half of the year was amplified by a surge in gold imports (the favourite saving vehicle of Turkish households in uncertain times) and was compounded by an increase in the interest costs of external debt. The current account deficit to GDP ratio reached 5.1% of GDP in the first quarter of 2020, 8.2% in the second quarter and 4.7% in the third quarter (Figure 1.3).

A deterioration in the financial account compounded the current account deficit. Capital outflows during the crisis were larger than in other emerging countries and lasted longer. Furthermore, foreign capital did not flow back as it did to other emerging markets. This was due, according to available indicators (including risk premia), to a weakening of investor confidence. At the same time, domestic non-financial businesses and banks continued to reduce their external debt as they were doing since the 2018 turmoil. This improved their balance sheets but reduced capital inflows (Figure 1.3 Panels C and D). Finally, the “net errors and omissions” item, which traditionally captures movements in Turkish savings parked abroad and tends to offset foreign financing shortfalls, moved this time in reverse direction. The resulting exchange-rate depreciation was sharp despite policymakers’ efforts to contain it (Figure 1.12 below).

Figure 1.3. The current and capital accounts have deteriorated

The recovery will be uneven and there are important risks

After a strong upturn in the third quarter of 2020, the recovery is expected to slowdown in the last quarter. Turkey’s output is projected to contract by around -0.2% in 2020. Uncertainty is high on the trajectory of the second wave of the pandemic, on its economic impact, and on future policy developments. Headwinds from the international environment, modest coverage of Turkey’s social safety net and low level of cash transfers, combined with firms’ and households’ increased debt levels, are projected to make the recovery more gradual than in previous post-shock upturns (Table 1.1).

The “New Economy Programme 2021-2023”, published at the end of September, had projected a slightly positive GDP growth of 0.3% in 2020, followed by 5.8% growth in 2021 and 5% growth in 2022 and 2023 (it had mentioned a risk variant for 2020 and 2021, with respectively a -1.5% GDP contraction on the first year and 3.7% growth on the second). The V-shaped baseline was obtained despite a tightening of the fiscal stance starting from the last quarter of 2020, and without additional monetary policy support, thanks to strong projected investment and export growth. Household consumption was expected to recover more gradually.

The New Economy Programme aimed at addressing a number of shortcomings in the business environment, in the entrepreuneurial eco-system, in Turkey’s digital skills and in the flexibility of the labour market. It sought to foster e-trade, to attract more foreign direct investment and to enhance environmental sustainability - specifically by converging with the European Union’s Green Deal. It aimed at increasing Turkey’s share in global value chains. At the same time, it stated that public procurement, the tax system, government-owned financial institutions and business incentives would be mobilised “to reduce import dependence and the imported content of domestic production”. Associated with the trade protection measures introduced during the COVID-19 crisis (which increased Turkish businesses’ cost of participation in global value chains - Dusundere and Koyuncu, 2020 and Akman, 2020) this commitment to reducing import dependence may conflict with the stated goal of deeper international integration of the Turkish economy. These policies can back domestic production and employment in the short-term, but they risk eroding the momentum of integration in global production networks, including in the European single market (Irwin, 2020).

New financial policy measures were introduced along the New Economy Programme. They relaxed partly the restrictions imposed on the operation of financial markets during the COVID-19 shock. First, the “asset ratio” for banks, which compelled them to expand their credits and investments in government securities, was scaled down (subsequently, following additional economic policy measures in November, its phasing out was decided from end-2020). Second, the exchange-transactions tax which penalised currency conversions was reduced. Third, the withholding tax on bank deposits was curtailed. Finally, the regulatory cap which restricted bank’s swap agreements with foreign counterparts was partly relaxed (see Box 1.5 below for more details). These measures were seen as positive steps by domestic and international investors, towards a more conventional operation of financial markets.

	2016	2017	2018	2019	Projections
	Current prices (TRY billion)				2020	2021	2022
Gross domestic product (GDP)¹	2,626.6	7.5	3.0	0.9	-0.2	2.6	3.5
Private consumption	1,560.4	5.8	0.7	1.6	0.8	3.9	5.7
Government consumption	387.0	5.4	6.5	4.3	2.7	2.1	0.1
Gross fixed capital formation	764.5	8.3	-0.3	-12.4	5.6	2.6	3.8
Stockbuilding²	-28.6	1.0	-2.9	0.1	6.2	-0.1	0.0
Total domestic demand	2,683.3	7.4	-1.6	-2.1	8.7	3.0	4.1
Exports of goods and services	606.3	12.4	9.0	4.9	-19.1	7.6	7.4
Imports of goods and services	663.1	10.6	-6.4	-5.3	7.7	9.3	8.8
Net exports²	-56.77	0.2	4.2	3.2	-8.5	-0.9	-0.8
Other indicators (growth rates, unless specified)
GDP deflator		11.0	16.5	13.9	12.6	12.1	9.9
Consumer price index³		11.1	16.3	15.2	12.2	12.0	10.0
Core inflation index4		10.1	16.5	13.4	11.0	12.0	10.0
Unemployment rate (% of labour force)		10.9	11.0	13.7	13.2	13.7	14.5
Current account balance (% of GDP)		-4.7	-2.1	1.2	-4.7	-4.6	-4.8

1. Based on working-day adjusted series.

2. Contribution to changes in GDP. Stockbuilding includes statistical discrepancy.

3. Based on yearly average.

4. Consumer price index excluding energy, food, non-alcoholic beverages, alcohol, tobacco and gold.

Source: OECD (2020), OECD Economic Outlook: Statistics and Projections (database).

Macroeconomic developments ahead will be highly sensitive to the sentiment of domestic and international investors concerning the quality and predictability of fiscal, monetary and financial policy frameworks. The borrowing needs of businesses and households, and, increasingly, of the public sector increase vulnerability to any adverse developments in risk premia and exchange rates.

External funding needs will encompass the financing of the current account, the rolling-over of maturing debt and the need to offset capital outflows. Direct financing needs (net of capital movements) are projected to reach 29.2% of GDP between October 2020 and October 2021. The renewal of maturing trade credits (estimated at USD 54.2 billion) and of the deposits of non-residents (USD 68.7 billion) should be smooth, but rolling-over bank, non-financial business and government debt (USD 58.1 billion) could be costly and demanding. Supportive international financial conditions are expected to facilitae external financing in the short-term, absent new tensions, but high risk premia will put pressure on the long-term sustainability of external debt (as discussed in more detail below).

The macroeconomic outlook is exposed to geopolitical risks. Turkey depends strongly on external trade and on value chain interactions with trade partners, which expose the economy to both downside and upside risks from geo-political developments. Interactions and relations with the EU (48% of Turkish exports in 2019), Near- and Middle-Eastern countries (19% of Turkish exports), the United States (5%%) and Russia (2.2%) are implicated. The opportunities arising from the restructuring of global value chains may be affected. On the other hand, improvements in co-operation prospects with the EU, UK, US and region’s countries could generate new opportunities for Turkish businesses. This applies in particular to the modernisation of the customs union agreement with the EU (Adar et al., 2020).

The Brexit process will have noticeable implications for Turkey, as it is a large exporter to the UK (6 % of Turkish merchandise exports in 2019). A Trade Working Group between the two countries is working on a Free Trade Agreement (FTA) to preserve the existing preferential trade conditions to the extent possible. Without such an FTA, key exports such as automotive, machinery, electronics -about 75% of all Turkish exports to the UK- would face tariff increases of 2 to 18%.

Turkey hosts the largest refugee population in the OECD (3.6 million) and this group is particularly affected by the COVID-19 crisis. They face higher health risks due to their living conditions (Deutsche Welle, 2020a). They are also estimated to have faced large employment losses as the majority work informally (Euronews, 2020b). Turkish authorities face therefore additional public health, social and fiscal challenges. Further refugee inflows may occur. Defense- and security-related spending is large, requiring its integration in the medium-term public finance framework. There are also, regrettably, natural disaster risks in the background. Table 1.2 outlines some exceptional events which could lead to additional changes in the outlook. A section below on well-being and social cohesion discusses some of them in more detail.

Events	Implications
A worsening of the COVID-19 pandemic abroad and at home	Further depression of external and domestic demand. Additional contraction of employment and incomes. Fiscal dilemmas between competing health spending, social solidarity and economic support programmes could worsen.
A negative confidence shock due to fiscal, monetary or financial policy uncertainties	Additional TRY depreciation, increasing risk premia, adverse capital movements and domestic financial contagion.
A severe worsening of geo-political tensions involving Turkey	Contraction and reversals in value-chain partnerships of Turkish businesses with selected countries. Negative impacts from trade and investment restrictions.
A major earthquake	Human tolls and crisis management challenges. Important losses of property and production capacity.

Policy priorities for containing the pandemic and supporting the recovery

Containing and managing the second wave of the pandemic is obviously key for economic recovery. OECD cross-country insights confirm that good policies pay and various model simulations indicate that, even in the absence of a general application of a vaccine, additional contagions can be reduced. After long-lasting solicitations, the resilience of the hospital system and of health professionals became a challenge in the second wave. Medical associations speak of growing tensions (Ankara Tabip Odasi, 2020), and according to their estimations (not confirmed by the Ministry of Health) there has been a rise in the number of health professionals withdrawing, resigning or on long-term sick leave (BBC News, 2020b). The Ministry of Health declared at the end of October that future resignations of public health personnel will not be accepted (Turkish Medical Association, 2020b) and announced the recruitment of 12.000 additional health professionals to support the existing staff and infrastructure. The testing, tracing and tracking system continues to function intensely (with more than 19.000 teams working full-time throughout the country) but there are concerns about its being overwhelmed by the resurgence of cases. Further efforts will be needed to preserve the preparedness and capacities of the public health infrastructure.

The “return to normal life” measures should be backed with stricter enforcement of physical distancing. Policymakers gained precious experience in fine-tuning lockdowns, selectively confining vulnerable groups, and isolating clusters. The results obtained so far should be re-assessed to identify the most effective procedures. As in all OECD countries, special attention should continue to be paid to the quality and accuracy of tests (OECD, 2020d). The number of cases and fatalities should be monitored and communicated according to international standards.

While a one-size-fits-all support strategy was justified during the first phase of confinements, policy support should now be adapted to the varying conditions of sectors, workers, households, and companies in the second wave. The economy will need to operate under partial confinement for some time as long as an effective vaccine is not widely applied. The reallocation of workers and capital resources to viable activities should be facilitated. Measures which postpone the liquidity strains in the business sector as a whole should be replaced gradually with supports to the post-shock investment capacity of promising firms and activities. The recommended priorities for Turkey’s support policies in the short run include:

Firms and workers in viable activities prevented from operating normally should continue to be supported. This should notably include the large enterprises with high fixed costs in the tourism, hospitality and entertainment sectors, which are crucial for Turkey. All firms receiving public support should be encouraged to prepare for post-pandemic economic conditions, including through re-training of workers and greening of activities.
The gap in employment-related social protection between formal and informal workers should be reduced. For vulnerable families who are not covered by employment-related protections, temporary but predictable allowances rather than irregular one-off transfers should be put in place.
Part of the subsidised and guaranteed loans to households and firms can be replaced with targeted and temporary transfers. For example, the one-off subsidy of TRY 1,000 to the 6 million households at risk of poverty during the COVID-19 crisis can be converted into a temporary but recurrent allowance for a limited period.
For young workers and graduates joining the labour market, further apprenticeship and internship programmes adapted to the post-COVID world should be put in place. Enterprises which benefit from government aids should be encouraged to implement such programmes. A temporary exemption of employer and employee social security costs could be granted to all young workers (in addition to the already existing “easy employer” scheme, which cuts social security contributions for firms employing young workers for less than 10 days per month).
Additional policy measures should ensure that working-age recipients of unemployment benefits and other social help actively look for jobs, and participate effectively to the re-training programmes on offer.

Strengthening macroeconomic fundamentals after the shock

Re-balancing demand and securing external sustainability

To shift to a sustainable growth path after the COVID-19 shock, Turkish economy needs to address its central structural imbalance. Growth is excessively driven by domestic consumption. Every time the economy operates closer to full employment, the current account deficit widens. The resulting dependence on foreign savings has entrenched under generally benign international funding conditions after the global financial crisis. Dynamic growth of domestic consumption under such circumstances typically fuels domestic price pressures, feeding into inflation inertia, triggering episodes of real exchange rate appreciation, and weakening external competitiveness. There have been periodical corrections through balance of payment strains, and related exchange-rate depreciation shocks, most recently in 2018, but they have not delivered durable adjustments. The impact of the 2020 depreciation remains still uncertain.

Policymakers have tried to re-balance the economy periodically but could not surmount the underyling structural challenge. Policy measures aimed at curbing household consumption, lifting-up household savings, and stimulating exports started to pay off (Figure 1.4). However, the re-orientation of the supply side of the economy towards exports has remained too slow and the aggregate supply potential could not expand at a pace fast enough to absorb the trend increase in the labour force (Figure 1.4 Panel D). This dilemma impelled policymakers to periodically revert to domestic demand stimulation, as they have done again after the 2018 shock and during the COVID-19 crisis.

This growth pattern has led to a steady deterioration in Turkey’s net international investment position (with a pause between 2018 and 2020, due to the cyclical impact of growth shocks and to the deleveraging efforts of the private sector). Beyond cyclical effects, the gross external debt stock is on an upward trend. Absent structural change, the external debt-to-GDP ratio will remain a concern for the sustainability of growth (Figure 1.5). As discussed in the thematic chapter, improving productivity and international competitiveness will be crucial for reverting to a sustainable growth trajectory. Fuller use can then be made of the economy’s resources, more and better jobs can be created, and people’s living standards can be raised without falling into stop-and-go cycles.

External sustainability can also be improved by reducing international investors’ risk perceptions. Turkey’s risk premium has risen since the onset of the COVID-19 crisis from an already high level - before declining at the end of 2020. While it had improved thanks to fundamental institutional reforms in the 2000s, permitting an outstanding decline in the economy’s funding costs (Gönenç et al., 2010), risk premia had increased again in the 2010s as policy and institutional uncertainties augmented (Box 1.2). In mid-October 2020, Turkey’s 10-year government borrowing costs in USD reached 6.8% against 5.2% for South Africa, 3.6% for Brasil, 2.8% for Poland and 2.7% for Mexico. Reducing policy and institutional uncertainties would ease investors’ risk perceptions, lessen external funding costs, stimulate non-debt capital inflows and enhance external debt sustainability (Box 1.2).

Figure 1.4. Re-balancing has made progress, but with slower growth and job creation

The discovery of natural gas reserves on Turkey’s Black Sea coast in August 2020 (estimated at 400 cubic meters) may positively affect structural external balances - independently from events in the Eastern Mediterranean. Yearly energy imports equal roughly the structural trade deficit. Natural gas imports gravitate around 45 billion cubic metres, corresponding to nearly 2% of annual GDP (depending on energy prices and on the business cycle). The contribution of the new gas reserve would depend on the pace of exploitation. The authorities estimate that exploitation can start in 2023 and that the energy bill can be reduced by 0.3-0.4% of GDP starting from that year. There are reports that additional reserves may be discovered.

Figure 1.5. External debt is expanding strongly

Box 1.2. Curbing the high risk premia

Turkey’s risk premia and external financing costs are very high in international comparison, for both credits and equities (Figure 1.6).

GDP growth, price stability, and the quality of governance institutions are key determinants of Turkey’s risk premia (Gül, 2020a). Technological progress in the business sector helps reduce the risk premium as investigated in earlier OECD Surveys (OECD, 2018; Özmen, 2019). Statistical analyses confirm that high risk premia pass through to lending rates and capital costs (Gül, 2020b).

High risk premia reflect on stock prices, by increasing the discount rate of investors. This is one of the drivers of the differentiation of the price/earnings ratios of listed firms. Panel B of Figure 1.6 provides a proxy of international differences in equity capital costs. In October 2020, the top 100 Turkish firms were trading at a 54% discount from emerging market peers, hinting at very high risk premia (Oyak Yatirim, 2020).

If Turkey could reduce its risk premia to the levels observed in the 2000s, external and internal debt sustainability would improve (Figure 1.5, Panel E and Figure 1.10). More stable funding costs would reduce the risks of balance of payment crises, decrease credit and equity capital costs, and stimulate investment and growth.

Figure 1.6. Risk premia are excessive and should be reduced

Moving to a more transparent and predictable fiscal framework

Turkey went into the COVID-19 crisis with a public deficit of 2.9% of GDP in 2019 and, despite the low level of public debt, extensive off-balance sheet commitments (Figure 1.9). This resulted from the massive government stimulus provided in 2019. At the beginning of 2020, staff expenditures had significantly grown due to job creation in the public sector. Higher borrowing costs had also lifted interest expenditures. In contrast, COVID-19-related on-budget costs had remained relatively limited in the first wave of the pandemic. According to the IMF Fiscal Monitor database, COVID-19-related spending and foregone revenues amounted to about 0.2% of GDP by mid-June 2020. The government has estimated these direct budget costs at 0.7% of GDP by the end of July (including the costs of the short-term working arrangement, the unpaid leave scheme, the additional unemployment insurance payments, and the one-off social support for the families in need).

The mainstay of Turkey’s COVID-19 support policies in the first wave was quasi-fiscal, not directly affecting the net lending of the government (see Box 1.1). Public bank loans and government loan guarantees, and, to a smaller extent, equity injections in financial and non-financial firms formed the backbone of government actions (Figure 1.7). Such ‘below-the-line’ supports amounted to 9.1% of GDP in the first five months of 2020 according to the IMF Fiscal Monitor database, going well beyond the emerging market ‘below-the-line’ average of 2% of GDP.

This distinct support system had distinct impacts on Turkey’s public finances, business and household balance sheets and on the financial system as a whole:

Ultimate impact on public finances. A sizeable share of the quasi-fiscal support offered during the first wave of the COVID-19 shock may turn into explicit fiscal costs. This is expected to result from the social security debt accumulated in the health system (Ministry of Development, 2018), which increased during the pandemic, and from loan defaults and calls on government guarantees. Even if Turkish banks entered the pandemic with, in principle, robust capital structures (Figure 1.8), the system’s non-performing loan (NPL) ratio increased from below 3% in early 2018 to 5.3% in January 2020 (a still low level given the severity of the 2018 shock in international comparison - Ari et al., 2020). It then declined to 4.1% by the end of August 2020, reflecting the fast expansion of new loans, the re-scheduling of existing loans under policy guidance, and the relaxation of loan classification methodologies along international recommendations (there is international consensus on the need to avoid classifying loans as non-performing after the standard 90 days delinquency during the pandemic). In its Financial Stability Report in November 2020, the central bank documented that while non-performing loan ratios improved in Turkish banks between 2019 and 2020 as a result of these restructurings and reclassifications, “loans under close scrutiny” account for around 10% of credit portfolios (CBRT, 2020b). Traditionally, 15% of these loans tend to turn non-performing but this proportion may worsen under demanding circumstances.
The loan classifications that the banking regulator (BRSA) has been implementing since 2019 and according to the latest international standards (IFRS 9) will permit a more refined monitoring of loan quality. The BRSA and The Banks Association of Turkey (TBA) are publishing detailed financial information on bank balance sheets on their websites (on a sectoral basis on the BRSA website and on individual banks on the TBA website) and these reports, with the help of internationally comparable classifications, will enable more refined data driven analyses of bank balance sheets by third-parties in the future. The quarterly external audit reports are also publicly available. Nevertheless there were qualms about the asset quality of some Turkish banks before the pandemic (IMF, 2019; European Bank for Reconstruction and Development, 2020). They were related to the suspected evergreening of bad loans in recent years (IMF, 2019).
The challenge concerning loan quality was amplified after the COVID-19 shock. The policy-stimulated credit growth in 2020, due to its exceptional pace, is expected to have reduced loan quality. The volume of public and private credits increased sharply between March and August 2020 (Figure 1.7). The guarantee provision capacity of the Credit Guarantee Fund was doubled in March 2020 (from TRY 250 billion to TRY 500 billion) and actual loan guarantees increased by 52% between June 2019 and June 2020. A significant share of these loans and guarantees were granted to households, firms and self-employed under financial constraints, which may be expected to continue to face strains during the second wave of the pandemic. While the budgetary impact of the loans extended by the Credit Guarantee Fund is limited to 10% of the outstanding loan amount, and both public and private banks classify and provision for their risky loans under the same standards, current indicators of loan quality, including the non-performing loan ratios, may fall short of highlighting sizeable future contingencies. The potential cost of loan defaults to public banks for public finances should be estimated, including under adverse scenarios.
The recommended Fiscal Policy Report can present this information. The banking regulator as well as independent third-party analysts can contribute to prospective analyses. OECD recommends the publication of asset quality reviews for individual banks and for the banking system as a whole (OECD, 2020n; European Banking Authority, 2020). Turkey is one of the few OECD countries not releasing stress test results for individual banks, out of concern for undue market impacts under limited financial literacy. Even though there is no general requirement with respect to Basel standards on publishing individual banks’ stress tests, disclosures can increase domestic and international confidence in the resilience of the banking sector (BIS, 2018). Cross-country research suggests that such disclosures are welfare-enhancing (OECD, 2020n).
Faced with macroeconomic sustainability concerns in markets, Turkish policymakers scaled down their quasi-fiscal activism from late August. Loan growth moderated, but stayed above historical trends until the very end of the year (Figure 1. 7). Loan conditions were tightened. Interest rates on public banks’ housing credits were, for example, raised from 8.4% in June to 11.3% by mid-summer and to 16-24% by the end of September. Rates on so-called “emergency loans” for households increased from around 15% in June to between 20-30% at the end of September. The banking regulator reduced its regulatory “asset ratio” in two steps, in August and September, to reduce its expansionary impact and, ultimately, phased it out from 31 December 2020.

Figure 1.7. Credit growth was exceptional after the COVID-19 shock

Figure 1.8. Banks have weathered the 2018 strains but will come under pressure after the pandemic

Debt burdens for businesses and households. Despite starting from a comparatively moderate level in international comparison, total debt accumulation since the onset of the pandemic amplified the risks of debt overhang in many businesses, and of excessive debt leverage in many households. Business debt as a share of GDP had soared before the pandemic, from 35% in 2009 to 69% in 2018 – one of the sharpest increases in the world. It settled at 66% in 2019, after active de-leveraging by businesses following the 2018 financial turmoil and mounted again through 2020. In keeping with the balance sheet analyses of the 2018 OECD Economic Survey of Turkey, young start-ups and medium-sized firms should face the highest risks of debt overhang. Industrial investment will suffer, notably in capital-intensive activities with high digitalisation needs such as tourism (Dünya, 2020) and retail trade. Business bankruptcies, which were adjourned between March and June 2020 will unavoidably grow. They increased by 10% over a year ago in August and an international study projected them to increase by more than 30% in 2021 - alongside a cross-country surge of 35% (KPGM, 2020). Turkey’s high risk premia is affecting adversely the feasibility and cost of financial restructurings.
Household debt was, at first sight, at benign levels in international comparison before the COVID-19 shock – at around 15% of GDP. It will increase during the pandemic as a result of loan-centred supports. It already increased by 33% between January 2019 and September 2020. The OECD Secretariat estimates that it may have reached 20% of GDP by the end of 2020 and a private forecaster projects it at 25% in 2022 (Trading Economics, 2020). This pace of expansion of credits is a source of risk for their quality (Alessi and Detken, 2018). The allocation of credits between different types of households will bear on their macroeconomic and social impact. Low-income households (for whom credits are the main source of income replacement to finance basic needs) will be constrained by excessive leverage. In contrast, households sheltered by social safety nets can use the subsidised loan packages for more discretionary purchases and can continue to borrow.
Systemic impacts on the financial system. The underlying developments in the financial system were accelerated by the COVID-19 shock. The share of government-owned financial institutions expanded, furthering a development that started in 2018. Guidances and regulations related to capital allocation, including those introduced as macro-prudential tools, have expanded. This included a constraining “asset ratio” for banks which penalised them if their pace of credit extension and security purchases fell below targeted rates (BRSA, 2020a). In the context of the monetary and financial policies introduced from November 2020 this regulation is repealed from 31 December 2020.
Banking has a central role in the Turkish economy (the correlation between credits and the business cycle is the highest among all countries reviewed by the Institute of International Finance in 2019). Fundamental reforms during the 2000s made commercial banks competitive, well capitalised and well regulated. However, one structural flaw was the tendency of commercial banks to engage in pro-cyclical lending as in other OECD countries (Huizinga and Laeven, 2019; Çolak et al., 2019). Whereas Turkey’s public banks are subject to the same legislation as private banks, and are in principle run under the same corporate governance rules, they undertook active countercyclical policies after the 2018 financial turmoil and during the pandemic. This has been visible in the divergence of their lending behaviour from private commercial banks during these downturns (Figure 1.7).
The establishment of a Sovereign Wealth Fund (Türkiye Varlık Fonu - TVF), with the aim of “providing resources for Turkey’s strategic investments” mirrors the same approach in equity financing (Box 1.3). In September 2020, the President also announced an intention to use the retirement savings accumulated in the 2nd pillar pension system (BES, worth approximately 3.5% of GDP as of September 2020 ) “as long-term and low-cost funding sources for the real economy”. No further details were made public. The BES system is currently managed by a competitive pension fund management industry (OECD, 2019c).
Government-owned banks generated 72% of the net credit increase in 2019, and more than 60% in the first half of 2020. Their weight in financial intermediation raises new challenges. While banking regulations are line with international good practices, and Turkey complies with Basel rules, extensive reliance on public banks raises risks. A recent analysis of government-owned banks’ lending found that it is strongly affected by Turkey’s national political cycle as well as local political circumstances (Bircan and Saka, 2019). Once the most acute phase of the COVID-19 pandemic is over, a transparent environment should be restored between different types of financial institutions. Public banks’ corporate governance practices, their competiton conditions with private banks, and the financing of their public service obligations should be closely examined in the light of international good practices (OECD, 2015c). Banking regulators should involve the Turkish Competitiion Authority to ensure a level playing field between public and private banks - as well as between public and private borrowers in access to finance. Such efforts would improve the pricing of risks and the efficiency of credit allocation.

Quasi-fiscal channels helped minimise the burden of the pandemic on public finances in the first wave, facilitated the distribution of liquidities, and rendered part of the transfers reimbursable. At the same time, the transparency of the support package, its targeting to businesses and households most in need, and its consolidation in a coherent macroeconomic framework was made more difficult. To support the recovery:

Fiscal policy should replace concessional credits to eligible households and businesses with reduced prospects to reimburse their loans, by direct temporary transfers. Fiscal room is available for such a re-balancing of support channels.
Fiscal tightening should resume only gradually, once the recovery is firmly underway.
The contingent liabilities that public bank loans and government loan guarantees raise for public finances (not only the guarantees underwritten by the Treasury) should be transparently gauged.
As long as the pandemic is not under control, all room available for fiscal support should be preserved for helping the health system and the households and businesses in need. Lesser priority plans should be postponed, to preserve room for rapid fiscal response to changing needs.

Figure 1.9. The fiscal stance loosened considerably already before the COVID-19 shock

Box 1.3. The Sovereign Wealth Fund (TVF)

Turkey’s Sovereign Wealth Fund (TVF) was created with a special law in 2016. It became part of the Presidency in 2018, the President becoming its Chairman of the Board and the Minister of Treasury and Finance its Deputy Chairman. Its board composition, its special legal status and its strategic mandate make it a unique entity. It aims at “developing and increasing the value of Turkey’s strategic assets, at providing equity for Turkey’s strategic investments, at financing large infrastructural projects, at deepening the local capital markets, at stimulating employment, and at supporting Turkey’s international economic objectives” (TVF, 2020). It is expected to “help reduce Turkey’s chronic current account deficit by investing in petrochemical, mining and energy sectors”.

As of October 2020 the Fund was led by Mr.Z. Sonmez, the former Turkey head of Malaysia’s national wealth fund Khazanah, who described TVF as “an Asian style asset-based development fund inspired by Singapore’s Temasek and Malaysia’s Khazanah”.

A large set of government assets were transferred to TVF. Its portfolio comprises listed and non-listed firms in various sectors, including financial services, energy and mining, transportation and logistics, technology and telecommunications, agriculture and food - including Turkish Airlines, Turkish Petroleum, Ziraat Bank and Turkish Post. According to its consolidated financial statements at the end of 2019 its total assets reached TRY 1.46 trillion (US $ 245 billion at that date) and its net equity TRY 234.54 billion (USD 39 billion).

TVF has access to a variety of funding sources, including cash and assets that may be transferred from other public institutions; dividend, rental and royalty income from the assets it owns; and direct funding from local and international financial markets.

It was granted immunity from a range of laws and regulations including the Law on the Court of Accounts (it is not audited by the Court of Accounts) and the Law on the Protection of Competition. This could reduce competition in the markets and activities where TVF intervenes. It is also exempt from certain taxes and charges including the stamp duty, income and corporate taxes, tax deductions and from the fees of the Borsa Istanbul. Under Turkey’s Banking Law (No. 5411) the loans to be made available to risk groups defined in the Banking Law cannot exceed twenty-five percent of their own equities. However, according to an amendment in this Law in February 2020, TVF is exempted from this limitation as it is not included in the designated risk groups.

TVF’s 2018 and 2019 financial statements were audited according to International Financial Reporting Standards (IFRS). It is a member of the International Forum of Sovereign Wealth Funds and has committed to comply with its ‘Santiago Principles’ (SWF, 2008). TVF could also draw on OECD’s “Guidance on Sovereign Wealth Funds” (OECD, 2008). This guidance contains principles and safeguards to help countries with both such funds and those receiving their investments to facilitate their operation in a transparent and open framework.

Boosting long-term fiscal credibility

Once the recovery takes hold, the medium-to-long term sustainability of public finances should be improved. OECD’s public debt projections presented in Figure 1.10, which take into account the costs of the COVID-19 shock, show that the prudent ‘fiscal policy debt limit’ estimated at 35-40% of GDP (see below) is already breached, and will be difficult to restore in the period ahead. Under unchanged policies, the public debt/GDP ratio is projected to increase strongly. Ageing-related spending as a result of the closure of Turkey’s demographic window around 2025-2030 is expected to bear on debt dynamics.

Recent changes in the composition of government debt, including its lower average maturity and the higher share of floating rate and foreign currency borrowings increased vulnerability to adverse developments in exchange and interest rates (Ministry of Treasury and Finance, 2020). In November 2020, the authorities announced an intention to increase again the share of long-term Turkish Lira borrowings. The outlook may turn more straining if the contingent liabilities accumulated during the COVID-19 shock move on balance sheet. This risk is not taken into account in the projections of Figure 1.10. Public debt dynamics could in contrast improve if Turkey’s trend growth rate is lifted-up and if risk premia and real interest rates are reduced thanks to the reforms recommended in this Survey (Figure 1.10).

Turkey’s room for fiscal manoeuvre would increase if its status on financial markets were upgraded. This would create fiscal space in the event of a renewed worsening of the pandemic, and would allow time to strengthen the public finances once the recovery is on track. An estimate based on past responses of Turkey’s risk premia to alternative public debt trajectories suggests that the public debt-to-GDP ratio should stay below the 50-55% band in order to cope with persisting exchange and interest rate risks. Once the public debt ratio reaches the 30-40% band, the countercyclical impact of fiscal stimuli starts to weaken, due to adverse impacts on risk premia and market interest rates (fiscal policy debt limit as discussed in Özatay, 2019). Such thresholds will vary in the post-pandemic world as a result of changes in global public finance benchmarks, but these considerations should be taken into account in the long-term planning of fiscal policy.

The reduction of general government debt from around 76% of GDP in 2001 to around 38% in 2008 – which positively decoupled Turkey from several other OECD countries – was a major achievement of Turkish macroeconomic policy (OECD, 2010). It triggered a massive fall in Turkey’s risk premia and created much welcome room for countercyclical fiscal policy. This room was effectively utilised following the global financial crisis, and more aggressively in the recent period, both after the 2018 and COVID-19 shocks.

Strengthening fiscal institutions would help improve the management of public finances and market credibility in the current circumstances (Box 1.4). Four goals matter most:

General government accounts should be reported according to international national accounting standards. These should be used as the central planning and communication instrument of fiscal policy. Government accounts are currently published by different agencies (including the Ministry of Treasury and Finance, the Strategy and Budget Unit of the Presidency, and Turkstat) using the same basic data (from the General Directorate of Accounting of the Ministry of Finance) but along their specific methodologies. There are only slight differences between these methodologies and each one has its respective utility, but the planning and communication of fiscal policy should be unified around a common set of international national accounting standards. This would facilitate the timely generation of general government accounts, their international comparability and their monitoring and analysis on a cyclically-adjusted basis. Central budget outcomes (significantly narrower than general government outcomes) should be focused at principally for high frequency indicators.
A Fiscal Policy Report, on the model of the Central Bank’s Inflation and Financial Stability reports, based on quarterly general government accounts, should review the totality of the above-the-line and below-the-line public revenues and expenditures, and below-the-line contingent liabilities.
Once the exceptional public finance conditions of the COVID-19 shock are behind, a fiscal rule should be re-introduced under the surveillance of an independent Fiscal Council, as in many other OECD countries. The rule developed in 2010 (but then not implemented) remains well adapted to Turkey’s circumstances (see Box 1.4).
A tax reform is compelling on both economic and social grounds. A key priority should be reducing labour taxes as discussed later in this chapter. Social protection should be financed from more employment-friendly sources. The recurrent tax amnesties should be discontinued. The recent digital taxes should be re-examined in the light of ongoing international co-operation (OECD, 2019h).

Box 1.4. Upgrading fiscal institutions

Building confidence in the long-term sustainability of public finances is essential. This should implicate both general government finances and quasi-fiscal and contingent liabilities. The 2005 Law on Public Financial Management and Control (Law 5018) should be fully enforced to this effect. This legislation defined a comprehensive fiscal policy framework (Yilmaz and Tosun, 2010). It prescribed a programme-based spending framework, area specific funding ceilings, performance benchmarks, and a good financial control and audit system. It vested the Court of Accounts and the Parliamentary Budget and Planning Commission with the task of monitoring the actual fiscal position, including off-budget liabilities.

Three streams of off-budget liabilities deserve special attention in Turkey:

The contingent liabilities of public financial institutions, including government-owned banks (Ziraat, Halkbank and Vakifbank), the Eximbank, the Development and Investment Bank, the Sovereign Wealth Fund and the Credit Guarantee Fund. The liabilities of these institutions expanded strongly following the August 2018 financial turmoil and the COVID-19 crisis.
The contingent liabilities of public-private partnerships (PPPs). Turkish PPPs reached the highest average investment size per project among all emerging countries – they are estimated at almost USD 600 million per project by the World Bank. On the other hand, according to the Turkish Presidency Strategy and Budget Unit database, the average investment size of PPPs is slightly above USD 300 million. More than 250 PPP projects in a wide range of activities were operational in 2020 (Table 1.3). Yearly disbursements for realised liabilities are published, but a prospective analysis of the obligations that may arise in the future is not available. These obligations will depend on commercial and financial contingencies – such as those currently experienced in airports. They may reach very high levels.

Sectors	Motorways	Airports	Energy	Health facilities	Ports	Industrial facilities	Marinas	Border Gates	Railways
Projects (Number)	41	18	97	18	23	2	18	23	1
Investment amount (US$ Billion)	23.6	19.1	18.2	10.5	2.1	1.5	1.3	0.6	0.3

* As of January 2020.

Source: Strategy and Budget Unit of the Presidency of the Republic.

Public pensions. Pensions are not included in public debt in a conventional sense. They represent nonetheless a substantial liability for public finances and should be properly assessed in efforts to secure their sustainability. The closure of Turkey’s demographic window around 2025-2030 is an important challenge (OECD, 2018a). Low average retirement ages, low contributor/beneficiary ratios, and the uncertainties concerning the indexation of future pension benefits create financial risks. Scenarios for long-term financial balances of public pensions should be regularly published in the recommended Fiscal Policy Report.

A formal fiscal rule adapted to Turkey’s circumstances, as was designed in 2010, should also be put on the agenda. This design was backed by the OECD at its inception (OECD, 2010). It was a “growth-based balance rule”, setting a ceiling for the annual general government deficit, which would be a function of: i) the general government deficit in the previous year; ii) the deviation of the previous year’s deficit from the long-term deficit target; and iii) the deviation of the GDP growth of the current year from the trend growth rate.

The objective of the 2010 fiscal rule was to maintain the public debt/GDP ratio at around 30% in the long-term. Policymakers were given three “windows” in the course of each year to adapt fiscal policies to the requirements of the rule: i) in the spring of the year t – 1, when preparing the medium-term economic framework of the year t; ii) in the fall of the year t – 1 when finalising the budget for the Parliament; and iii) in the spring of the year t, when the growth and fiscal outlook become more precise.

If a fiscal rule of this type, which is well-adapted to Turkey’s circumstances is implemented, an independent Fiscal Council can be vested with its monitoring and implementation as in other OECD countries.

Figure 1.10. High interest costs and ageing create pressures on public debt

General government debt, percent of GDP

Stabilising inflation and increasing monetary policy credibility

The COVID-19 shock amplified the longstanding challenges of Turkey’s monetary policy (OECD, 2018a). Inflation is high, stuck at a level well above the official target of 5% and its responsiveness to the cyclical position of the economy is low (Figure 1.11). In the face of the high output and employment cost of disinflation, and under the appeals of the executive authority, the central bank has long been perceived by international investors as prioritising growth and employment over price stability (Goldman Sachs, 2020; Citibank, 2020). Furthermore, during periods of decline in risk appetite in international markets, capital outflows tend to lead Turkish authorities to try to contain exchange rate depreciation through direct and indirect interventions. This tends to generate tensions with Turkey’s officialy open capital account, currency convertibility and floating exchange-rate regimes – compounding investor uncertainties. The COVID-19 shock has amplified this policy conundrum:

Increases in inflation. Inflation and inflation expectations augmented after the COVID-19 shock from an already high level (Figure 1.11). Additional price pressures resulted from exchange rate depreciation, cost increases in value chains, changes in work organisations, and, in certain markets such as housing and motor vehicles, from credit-fuelled demand. Inflation expectations picked up and remained significantly above the official inflation target as well as the official inflation projection (that the central bank asserts as “an interim target when inflation deviates significantly from target”- CBRT, 2019a). The central bank lifted its end-year inflation projection from 7.4% in April 2020 to 8.9% in July and 12.1% in October. Market expectations reached 12.5% in November, whereas actual inflation reached 14% in November and 14.6% in December, heralding a further worsening in expectations.
Monetary stimulus through new channels. In response to the COVID-19 shock, the Central Bank announced various “Measures Against the Economic and Financial Impacts of the Coronavirus” (Box 1.1). It slashed its policy rate to 8.25% in May and pulled down the real policy rate to negative territory (both on an ex-post and ex-ante basis). It launched quantitative supports as in other emerging countries (Benigno et al., 2020), offering additional liquidity windows for banks, larger rediscount facilities for businesses, and higher ceilings for government securities in its portfolio. As a result, the monetary base, the size of the central bank’s balance sheet and total money supply have all expanded (Figure 1.11, Panels E and F). By August, the expansion of money supply (M1) was the fastest among emerging countries and reached an annual increase of 70% -- against an emerging countries median of 11%. These developments increased uncertainties about the viability of the official inflation target.
Faced with an acceleration of capital outflows and exchange rate depreciation through Summer, the Central Bank took tightening steps. It started to tighten liquidity in August, raised the policy interest rate by 200 basis points to 10.25% at the end of September, and a further 475 basis points to 15% in mid-November. Between these two increases, it refrained from lifting up the policy rate directly, and tightened liquidity indirectly by offering funding through higher cost channels. This increased its effective funding rate to 13.40% by the end of October, but, despite this significant tightening, the divergence between official and effective monetary stances was interpreted by markets as a sign of political constraints to the independence of the Central Bank. These constraints had increased after legislative changes in 2019 which shortened the tenure of the top management of the Bank and facilitated conditions for its removal (actually permitting the removal of one governor in mid-2019). All in all, developments during the COVID-19 crisis increased market uncertainties over the future course of monetary policy, fuelled risk premia and accelerated exchange rate depreciation until November 2020. Against this backdrop, the appointment of a new governor in November 2020, with the Bank re-iterating price stability as its fundamental objective, associated with a consequential increase in the policy interest rate and its re-confirmation as the main channel of liquidity provision improved investor expectations. Risk premia and exchange rates eased.
Both policy and effective funding rates had stayed in negative territory in real terms (on an ex post as well as ex ante basis, as discounted by current and expected inflation) during most of 2020. They turned positive only in November. If expectations do not converge with the Bank’s inflation target and projections in the period ahead, and if risk premia and exchange rates are not durably appeased, the real rate would need to be lifted up further. It would need to be kept firmly and consistently in positive territory to regain credibility for monetary policy.
Capital flight and capital flow management. Several measures were taken in the past two years to reduce capital outflows and counter exchange-rate depreciation. These included taxation of foreign exchange operations, restrictions on foreign exchange derivatives, limits on foreign exchange operations of banks, and new regulations to curb “manipulative and abusive behaviour” in financial markets (Box 1.5). Public banks also intervened to help stabilise the currency, which increased their open position to around double the amount of the normally authorised regulatory limit in summer (38% vs. 20% in the first week of August). This was subsequently corrected by public banks increasing their foreign currency-denominated assets. These interventions created uncertainties on the degree of capital account openness, currency convertibility and exchange rate flexibility. Some of these measures started to be rolled back from late September (BRSA, 2020b).
Foreign exchange reserves shrank as a result of these interventions. Gross foreign currency reserves fell from USD 106 billion in March 2014 (11.3% of 2014 GDP), to USD 85 billion in March 2018, USD 54 billion in May 2020, and USD 40 billion in August 2020 (5.3% of 2019 GDP). Compliance with the standard reserve adequacy metrics of the IMF fell from the minimum required level of 100% in March 2014 to 67% in mid-May 2020 according to the official estimation of the IMF (IMF, 2020).
Net reserves -- foreign exchange assets minus liabilities- also declined. According to the CBRT definition, they fell from USD 36.8 billion in January 2020 to USD 23.3 billion in August 2020. According to a wider non-official definition (which also subtracts the short-term swap resources, assimilated to short-term debt) they fell, during the same period, from USD 17.8 billion in January 2020 to USD - 8.8 billion in June and to USD - 35.3 billion in August (a negative position) (Eğilmez, 2020a). There is market demand for more detailed net reserve gauges. This invites the publication of complementary indicators of the net reserve situation and active communication according to these market demands.
Swaps with foreign (central) and domestic (commercial) banks were indeed the main source of external funding for the central bank during 2020. They represented 63% of its gross reserves (including gold) at the end of June and 76% at the end of August. The authorities aim to establish additional international swap agreements to improve gross reserves. International swap transactions with other Central Banks are also used for supporting international trade with local currencies. Such agreements, or other international arrangements offering stable external funding options, would help, given sizable external liabilities and dependence on short-term funding.

All in all, monetary policy interventions in response to the COVID-19 shock backed economic activity, provided liquidity to the banking system, supported the exchange rate, helped to prevent a surge in corporate insolvencies and eased the financing of the Treasury. However, they also exacerbated uncertainties about the multiple objectives and instruments of the central bank. Exchange rate interventions by public banks – which are not publicly communicated – have amplified contingent liabilities for public finances.

This monetary policy framework has contributed to dollarisation. The dollarisation of households’ and businesses’ bank deposits, and large firms’ liabilities, is high in Turkey. While dollarisation shrank following 2000s’ reforms, as confidence built-up for the Turkish Lira, it rose again in the 2010s, passing, for bank deposits, above 40% in 2015 and 50% in early 2020. It further increased after the COVID-19 shock, boosted by negative real interest rates and uncertainties on the path of monetary policy, attaining 56% for household deposits in November. Dollarisation undermines the efficacy of monetary policy and adds to the volatility of the exchange rate (Estevão and Everaert, 2016). International research on its determinants find that curbing inflation and improving macroeconomic stability reduce it strongly. It also suggests that administrative measures to limit dollarisation (Turkish authorities took periodically such measures, although at a moderate intensity) increases the economy’s funding costs as long as fundamentals are not strengthened (Ergun et al., 2017).

Foreign investors have withdrawn from Turkish Lira-denominated financial markets in a more structural way. The share of foreign investors in the outstanding stock of government bonds fell from 20-25% between 2012-2018, to 10% in 2019, and to below 4% in July 2020, despite favourable global capital market conditions over much of this period. Exits from bond and equity markets reached respectively USD 7.6 billion and USD 5.8 billion in the first ten months of 2020.

International experience suggests that policy interventions and capital controls leading to such withdrawals may negatively affect long-term investment and growth (Andreasen et al., 2019). Recent research on vulnerable emerging economies suggests at the same time that well-confined capital flow and foreign exchange-rate measures, if implemented on a transparent and rule-based mode as prescribed by the OECD Code of Liberalisation of Capital Movements (OECD, 2020f), may have stabilising effects on output and employment (Adrian and Gopinath, 2020). Research also suggests that monetary policy rules responding to real exchange rate, asset price and credit spread developments, in addition to standard inflation and output gaps, may be effective in emerging economies (Mimir and Sunel, 2019).

To strengthen and consolidate market confidence, monetary policymakers should aim at:

Restoring domestic and international confidence in the independence of the Central Bank. Legislative measures reinforcing the inamovibility and extending the tenure of its management would send the strongest signal.
The real policy interest rate should be kept in positive territory as long as inflation and inflation expectations diverge from official projections and targets.
The Central Bank should spell out a strategy for rolling back, as soon as circumstances permit, the exceptional pandemic liquidity measures, as emphasised in its policy statements.
Capital flow management measures and exchange interventions should be resorted to only in exceptional circumstances and in accordance with the OECD Code of Liberalisation of Capital Movements.
Foreign reserves should be replenished as conditions allow. In addition to standard reports on their level, as currently available according to international standards, active communication by the central bank on various aspects of its reserve position according to the information needs of financial markets, including with the help of published research, would improve confidence.
Various public concerns about statistical methodology, data source and data quality issues related to inflation and monetary policy should be explicitly addressed (even when they may be misplaced). Central bank’s high-quality analysis and communication instruments such as the Inflation and Financial Stability Reports can be used for this purpose.

Figure 1.11. Monetary policy has failed to achieve the inflation target

Box 1.5. Capital flows and exchange rate measures

The Central Bank (CBRT) and the Banking Supervisory Authority (BRSA) implement several measures impacting on capital flows and the exchange rate. CBRT has a legal responsibility to steer the value of the national currency and has been very active in this area since the 2018 shock. Its official policy document states that “the Central Bank has no nominal or real exchange rate target” but “if the exchange rates deviate significantly from economic fundamentals and their movements permanently affect price stability or pose risks to financial stability, it will respond with the instruments at its disposal.”.

Exchange rate measures take place under Turkey’s fully liberalised capital account and floating exchange-rate regime, which have served the economy well over the past two decades. They permitted to bring in large amounts of foreign savings and provided an adjustment mechanism for macroeconomic shocks. However, free capital flows make also the domestic credit cycle, and, through it, the overall business cycle very sensitive to changes in global risk appetite and in investor sentiment concerning Turkey. Turkish policymakers have put in place, through the 2010s, macroprudential tools to temper this sensitivity. In the absence of tested international good practices, they developed, on a trial-and-error basis, an instrument set based on different provisions for different types of international bank liabilities (Kara, 2016).

These buffers proved broadly effective in de-synchronising capital inflows and GDP growth. However, their efficacy may have declined in the most recent period. The correlation between net capital inflows and GDP growth had attained a high coefficient of 0.54 between 1999 and 2010, then had fallen to as low as -0.17 between 2010 and 2016 and increased again to 0.22 between 2016 and 2019.

After the 2018 financial shock, as well as during the COVID-19 crisis, Turkey dealth with very volatile capital movements. These arose from both alterations in global risk appetite, as well as from increased Turkey-specific investor uncertainties and risk perceptions. They underpinned a significant trend deterioration in nominal and real exchange rates (Figure 1.12)

Figure 1.12. Nominal and real exchange rates have deteriorated before the recent uptick

Turkish authorities introduced several measures during this period to prevent further shorting of the lira and limit further volatility. These included:

Restrictions on foreign currency (FX) loans to corporates (May 2018): legal entities with FX liabilities of less than USD 15 million may not borrow in FX more than the sum of their FX income of the last three fiscal years, with some exceptions.
A limit on banks’ currency swap, forward and option transactions (where banks receive TRY at the maturity date) with non-resident partners at 10% their capital since 25 September 2020 (25% on 15 August 2018, 10% on 8 February 2020, 1% on 12 April 2020). On 11 November 2020, the limit for swaps, forward and option transactions where banks pay TRY at maturity was raised to up to 30% (depending on their remaining maturities) .
A limit on Turkish banks’ lira denominated transactions with non-resident financial institutions (5 May 2020) including repos, deposit facilities and loans, at 0.5% of banks’ equity, with some exceptions (the overdraft credit facility for foreign banks were exempted from this limitation).. On 27 November 2020, the mentioned ratio was increased to 2.5 % of banks' equity.
Mandatory repatriation and surrender requirements on FX export proceeds (September 2018). These must be repatriated generally within 180 days and at least 80% should be surrendered to a local bank in exchange for Turkish liras. On January 2020 the surrender requirement was dropped, but the repatriation requirement remained.
Prohibition of property, labor and service contracts denominated in foreign currency between Turkish residents.
A settlement delay of one day for FX purchases by individuals of more than USD 100 000 (May 2019). This provision was repealed on 8 December 2020.
A tax on foreign exchange sales by banks of 0.1% of the value of the transaction (May 2019). This tax was hiked to 0.2% on 7 December 2019 and to 1% on 24 May 2020. It was reduced to 0.2% on 30 September 2020.
Reserve requirements for banks, differentiated by currency, were adjusted several times across maturity brackets and liability types.
On 7 May 2020, a new regulation was published “to curb manipulative and misleading transactions in the financial markets”. Among other acts, the following are considered manipulative and misleading for the purposes of this regulation:
- providing false or misleading information on the supply, demand or price of a financial instrument.
- engaging in transactions that affect or may affect the price of a financial instrument through a deceptive mechanism or setup.
- conveying false or misleading information about a reference value, providing false or misleading inputs, or taking, knowingly, any manipulative behaviour affecting the calculation of reference values.
In May-July 2020, regulators enforced certain activity bans:
- On 7 May 2020, Turkish banks were prohibited from conducting foreign-currency trades involving the Turkish Lira with three banks, UBS Group, Citigroup, and BNP Paribas, reportedly due to the failures of these three banks to meet their TRY liabilities on time. The trading ban on the three banks was lifted on 11 May 2020.
- On 6 July 2020, the Istanbul Stock Exchange prohibited six foreign banks from short selling stocks on the Turkish stock exchange for up to three months. Barclays, Credit Suisse and Merrill Lynch (part of Bank of America) faced a three-month restriction, and Goldman Sachs Group, JPMorgan Chase, and Wood & Co. a one month restriction. The banks had failed to comply with a requirement to notify the authorities about their short-selling trades. Short sale restrictions for Goldman Sachs Group, JP Morgan Chase, and Wood & Co. ended on 6 August 2020, while those for the other three institutions ended on 6 October 2020.
On 18 April 2020, the banking regulator introduced a new “Asset Ratio” for banks (except development and investment banks and small banks), aiming at increasing their lending during the COVID-19 crisis. The ratio was adjusted a number of times and had substantial implications on banks’ lending and foreign exchange borrowing decisions in 2020. Although it was phased out from 31 December 2020, its implementation principles which played a major role are summarized below:
- The sum of a bank’s loans, 75% of its securities portfolio and 50% of its Central Bank swap balances must exceed the sum of its TRY deposits and 125% of foreign currency deposits. On 29 May 2020, the coefficient of “FX Deposit” was hiked to 175%. On 10 August 2020, BRSA stated that the Asset Ratio should not fall below 95% for the deposit banks and 75% for the participation banks at the end of each month. On 28 September 2020 the limits were lowered respectively to 90% and 70%.
- Some banks may not be able to meet the ratio by boosting lending, and may increase their holdings of government securities, or foreign exchange swaps with the CBRT. The ratio creates an incentive for banks to increase their FX swap transactions with the CBRT, which will boost the central bank’s gross FX reserves. On the other hand, in order to meet the target, banks may reduce the FX deposits of their clients, as these have a higher weight than TRY-denominated deposits in the asset ratio, thereby reducing the FX liquidity buffers.

Strengthening employment, job quality and social cohesion after the COVID-19 shock

The economic slowdown of the past three years culminating with the pandemic reversed the progress achieved in the labour market over the years towards better quality job creation and consequent gains in well-being and social cohesion. The shock has hit informal workers the hardest, as many work in contact-intensive services and may find it hard to apply physical distancing measures in their frequently sub-standard workplaces. They are also excluded from employment-related formal social safety nets. Young jobseekers were particularly affected by the contraction in net job creation.

A large portion of workers do not hold formal wage-earning jobs (Figure 1.14). The remarkable firm-to-firm and region-to-region labour mobility (Akgunduz et al., 2019) have not sufficed to allocate labour resources to higher productivity firms. The majority of workers are employed in micro-size informal or semi-formal activities. Despite a strong increase in women’s labour force participation, backed by myriad policy initiatives (which grew from 23% in 2007 to 34% in 2019, before declining to 32% in August 2020, against 70% for men) most women stay inactive, or work as unpaid family workers (Figure 1.15). There is a close link between informal work and family poverty across regions, skills and occupations of breadwinners. More than 15% of children live in this context in relative poverty, the third highest proportion in OECD (OECD, 2019f). Recent refugee inflows have amplified these disparities.

Informality arises from several factors, mostly related to labour regulations and labour costs. Turkey’s employment rules for both permanent and temporary workers are among OECD’s most rigid (Figure 1.16). Gross labour costs are inflated by high labour taxes, reflecting both the cost of the pension scheme (amounting to deferred wages) but also of a universal health insurance system partly funded by these taxes. Turkey boasts also one of the OECD’s highest minimum wage/median wage ratios, despite deep productivity differences between firms and regions. This ratio increased the most in Turkey in the entire OECD area over the last two decades. Informal and semi-formal firms that do not comply with regulations have flexible employment relations and low employment costs, which provide them with advantages over formal sector competitors. Large corporations, multinational firms and high-quality start-ups cannot benefit from such flexibility. The resulting frictions were identified in earlier OECD Surveys as a core impediment to efficient resource allocation in the Turkish business sector (OECD, 2016; Atabek et al., 2016).

The August 2018 and the COVID-19 shocks have amplified these labour market challenges:

Total number of jobs (formal and informal) contracted by 2.4 million workers between May 2019 and May 2020. The strong recovery in the third quarter of 2020 reduced net losses to 0.7 million between September 2019 and September 2020 - a decline of 2.6%. The employment rate, Turkey’s key indicator of labour market performance fell from 48% in April 2018, to 41.4% in May 2020, before partly recovering to 44.1% in September 2020 (Figure 1.13). Despite massive withdrawals from the labour force, the unemployment rate reached 12.7% in September 2020 and 24.3% for the 15-24 year-olds.
If men and women’s labour force participation had continued to increase along their earlier trajectories (i.e. without the last two years’ labour force withdrawals) unemployment would have reached higher levels. According to one estimation, it would have reached 19% in July 2020 (Tükel, 2020). The “broadly defined rate of unemployment” (according to US Bureau of Labour Statistics’ U-6 benchmark, which includes workers not looking for a job but are ready to work and the seasonal workers, therefore broadly correcting for temporary withdrawals from the labour force), rose from 19.5% in September 2019 to 22.9% in September 2020.The share of “youth neither in employment nor in education or training” increased from an already high 24% in May 2019 to 29.1% in September 2020. The size of this group, whose income and employment prospects are being eroded, is a severe challenge to Turkey’s well-being and social cohesion. It invites active policies to give young people greater opportunities on the labour market, through labour cost cuts, entrepreneurship and vocational education.
The two job retention schemes that Turkey used against the pandemic, i.e. the short-time work scheme and the government-paid furlough arrangement (18.3% and 7.6% respectively of total wage-earning employment in May 2020) have kept the level of employment above the level it would have fallen to without these buffers. A quarter of employment in summer was “retained” through these policy measures – a higher rate than in many other OECD countries (OECD, 2020g). The “number of workers in effective activity” reported by Turkstat fell from 26.2 million in January 2020 to 20.8 million in May (a 20.4%) decline), before improving to 25.7 million in September.
Remote (on-line) working diffused more rapidly than expected in information-intensive businesses and in the public sector, including a substantial shift to e-commerce in export activities and in retail trade (Turkonfed, 2020). On-line work is difficult in contact-intensive services, which are pervasive in Turkey. According to one estimate, 20% of existing jobs lend themselves to remote work, against 40% in advanced countries (OECD, 2020e). There are large differences between regions in this area: the potential ranges from 30% in Istanbul to 15% in Eastern provinces. The degree at which this potential was mobilised in different activities and regions requires further investigation.
The enforcement of employment rules may have weakened during the crisis. As the shock impacted the informal wage earners and the self-employed the most, the aggregate share of informal workers fell from 36% of all employed in September 2019 to 32.2% in September 2020. At the same time, semi-formal practices which help to circumvent regulations and legal contracts appear to have augmented. In asymmetric bargaining positions, a number of firms demanded wage cuts from their employees as an alternative to lower-paid furloughs. The enforcement of new health and safety rules against the pandemic (Ministry of Family, Labour and Social Services, 2020a) became a challenge in the high-informality economy (International Labour Organisation, 2020) Statistical evidence is practically impossible to collect in this area but there are many reports about their uneven enforcement between formal, semi-formal and informal firms. There were also many cases of good co-operation between firms and employees for adapting work modes to new circumstances. They are easier in sectors and firms with well-established social dialogue and collective negotiation practices (Avrupa Birliği Başkanlığı, 2020).
Despite the trend increase in women’s labour force participation mentioned above, Turkey’s gender gaps in labour markets (OECD, 2020) were adversely affected by the pandemic. Women’s labour force participation and employment rates had remained the lowest of OECD in Turkey, reflecting several factors including skill mismatches, undersupply of childcare services, inadequate maternity leave entitlements (notably in the informal sector) and cultural specificities (Figure 1.15). Unlike during the 2018 crisis, which caused larger employment losses among men than women (due to the collapse of construction jobs), the COVID-19 shock hit Turkish women harder than men, as they are over-represented in service sectors and in informal employment. Female jobs contracted by 5.2% between September 2019 and September 2020, against 1.4% for men. Women’s total work hours fell by 9.8% in September 2020 over a year ago, against 5% for men (average working hours of employed men and women decreased respectively by 2.1% and 2.2% between September 2019 and September 2020). During the lockdowns, women fulfilled the bulk of child and elderly care and contagion mitigation responsibilities, dealing with considerable additional strains.

OECD recommendations on employment policies during the pandemic apply fully to Turkey (OECD, 2020). As a general policy goal, the job-retention schemes in place (both the short-time work and the furlough schemes), implemented in the entire economy, should be converted to more selective arrangements to facilitate the shift of jobs from unviable to viable activities. All OECD countries are currently looking for ways to achieve such a fine-tuning of job protection programmes. For instance, firms can contribute more to the cost of hours not worked and to the re-training of workers during reduced work time. The ban on dismissals imposed in exchange of government-funded unpaid leaves can be gradually replaced, as proposed by a number of social partners. Firms benefitting from government aid can also be encouraged to create apprenticeships and internships for young workers. Temporary subsidies can support such inclusive initiatives.

Turkey’s long-delayed labour market reforms become more compelling after the COVID-19 shock, for three reasons. First, resuming net job creation, especially for low-skilled workers, become crucial for regaining the ground lost in social cohesion. Secondly, as informal workers are not covered by employment-related social protections, reforms facilitating transitions from informal and semi-formal to fully formal jobs become more urgent. Finally, well-functioning labour markets are a key requirement of global value chain operators re-organising and redistributing their activities and Turkey should not fall behind.

Policymakers should draw on the new OECD Job Strategy, which stresses emerging economies’ need to shift to formal labour markets (OECD, 2018b). The OECD strategy suggests that cutting non-wage labour costs, shifting part of the financial burden of social protection to sources other than social security contributions, making statutory minimum wages affordable for low-productivity firms, and modernising labour regulations for temporary as well as permanent contracts would stimulate job creation in the formal sector once the recovery takes hold. The mobilisation of the existing potential for increasing women’s labour force participation requires,among other factors, an increase in the availability and quality of early child care and education infrastructure.

Re-balancing the roles of statutory law and of enterprise-level negotiations for improving wage and work conditions has become more important (Boeri et al., 2019). A shift to adaptive negotiations would help all firms, including those with lower productivity, to comply with legislation, to escape informality and to upscale. This implies a wider adoption of bargaining between social partners than is currently the case in Turkey (where only 12% of workers participate in such processes). The pressures of the pandemic for the viability of firms and jobs could help re-fuel social dialogue, for both the modernisation of work modes at enterprise level and the adaptation of nationwide labour regulations. An “OECD Jobs Strategy Country Review” could help policymakers, employers and workers alike to discuss further a country-specific reform package (OECD, 2020i).

In the short-term, the special circumstances of the pandemic are not supportive of comprehensive structural reforms. Labour costs in the formal sector can nevertheless be alleviated with temporary or permanent cuts in social security contributions, by shifting part of their financing to more employment-friendly sources. The government is already implementing such cuts in certain regions and for certain special categories of workers. The schemes subsidising the social security contributions of persons aged 18-29 who set up their firm (this scheme supported around 125.000 entrepreneurs between mid-2018 and mid-2020) and the older and broader-based reductions of 5 percentage points in employers’ social security contributions for all wage-earners (applied since 2006) are estimated to have been helpful. A systematic assessment of the impact of these schemes by the Ministry of Family, Labour and Social Services is ongoing.

Figure 1.13. The COVID-19 shock worsened the already weakened labour market situation

In the pressing circumstances of the pandemic, policymakers could consider an additional exemption from employer and employee social security contributions for all workers below 25 for a temporary period. They may also consider a structural initiative when circumstances permit it, by durably reducing part of employer health insurance contributions of all workers as a first step in a broader employment-friendly reform of the financing of the health system. Appendix B offers an estimation of the approximate fiscal cost of such measures.

Figure 1.14. Unregistered employment varies across regions, sectors, firm sizes and education levels

Unregistered employment and employees, 15 years old and over

Figure 1.15. Labour force participation is low, particularly for women

Figure 1.16. Labour market reforms would stimulate job creation in the formal sector

Resuming progress in well-being and social cohesion

Turkey is among the OECD countries where different indicators of well-being improved more than average between 2010 and 2018 (Figure 1.17). This resulted from improvements in employment and household incomes. Total employment was of nearly 20 milllion in 2005, corresponding to 41% of the working age population. It soared to 28 million in 2019, mobilising 46% of the working age population in more productive sectors and higher quality jobs. According to Turkstat’s Life Satisfaction Survey, the “rate of satisfaction of workers with their earned income” rose from 19% in 2003 to 49 % in 2016 - before regressing to 43% in 2019.

Public finances have also improved since the 2000s, permitting to upgrade social services and, notably, allowing a path-breaking transition to universal health insurance. The share of citizens “reporting satisfaction with social services” increased from 40% in 2003 to 70% in 2013 - before weakening to 61% in 2019. Social transfers to the poor augmented. The rate of absolute poverty (the share of people living with less than USD 4.3 per day) fell from 30.3% in 2002 to 3.7% in 2010 and to 1.6% in 2015 (the latest year for which this data is available). Turkey continues to display nonetheless modest well-being standards in comparison to OECD benchmarks (Figure 1.17).

Figure 1.17. Well-being can be improved in several dimensions

The principal well-being gaps against OECD benchmarks concern household incomes and wealth, work-life balances and environmental conditions (OECD, 2020j). Turkey does better than OECD averages in social interactions and public political participation. Subjective life satisfaction, after improvements until mid-2010s, lost ground in the last 4 years. Beyond national averages, heterogeneities remain deep between households according to breadwinners’ labour market position (Figure 1.18).

Regional convergence of living standards

Gaps in living standards between regions are substantial and, following a welcome phase of catching-up in the 2000s, have stalled. Provinces with the lowest incomes per capita had grown more rapidly than others in the 2000s (Figure 1.20, Panels A and C). This momentum has subsequently lost steam (Figure 1.20, Panels B and D). Wealth inequalities remain currently one of the deepest in OECD (Figure 1.19).

Figure 1.18. Working-age people who are unemployed or in inactivity are more likely to be poor

Percent of population groups living in households reporting severe material deprivation

Figure 1.19. Income and wealth inequalities remain high

Figure 1.20. The convergence of living standards tended to stall

Social trust

Diversity in cultures and lifestyles is a genuine richness of Turkish society. At the same time, they form a more challenging environment for sustaining social trust than in other OECD countries. Self-perception of these differences may have increased (Kadir Has University, 2020). According to standard surveys, Turkey’s citizens hold currently the lowest level of “average trust in others” among all OECD countries (Algan, 2018).

Gender inequalities are not supportive of social trust. Violence against women is a source of special concern. According to the latest OECD indicator available, Turkish women report the highest proportion of physical or sexual violence from partners among OECD countries, at a rate of 38% (OECD, 2020k). The balance between paid vs. unpaid work by men and women is also the most uneven in OECD (OECD, 2020j). In contrast, various indicators of gender gaps among university graduates are low in OECD standards, confirming the strongly favourable impact of education.

The prospect of EU membership, which was a unifying aspiration through various sections of society when accession negotiations started in 2005, has since weakened. Support for EU accession declined from 70% in 2004 to 51% in 2019 (Kadir Has University, 2020).

Intentions to emigrate are high in certain fringes in the population and the actual emigration rates increased (Türkstat, 2020). A recent analysis found that one additional year of schooling increases the probability of reporting an intention to emigrate by 24 percentage points (Geverek et al., 2019). The Presidential Programme 2020 identifies this trend and declares that investigations will be undertaken to apprehend the factors driving brain drain, with a view to encourage reverse returns.

The region’s special challenges

Hosting 3.6 million refugees (4.3% of total population at the end of 2019) is a major challenge for Turkey (United Nations, 2019). Refugees represent a comparable proportion of the working age population. A minority has been cleared for formal labour market participation, but the great majority work informally, amplifying informality in regions where the refugee population is large. A survey by the Turkish Confederation of Trade Unions found that this resulted in larger proportions of SMEs’ resorting to informal employment, as domestic workers start to accept such jobs because of increased competition for work (Türk-Iş, 2019). One quantitative analysis found negative effects from refugees’ presence on low-skilled workers’ employment and wage levels (Bağir, 2018). A follow-up investigation confirmed that refugee inflows reduce aggregate demand for low-skilled labour, but also increase demand for skilled labour (Ceritoğlu et al., 2017). They seem to incite local families to lift up school attendance by their children as well as the academic commitment of these children – an indirect positive impact on human capital (Tümen, 2019).

People under temporary protection are granted free access to health and education services at pre-primary, primary, secondary and tertiary levels - a generous and inclusive policy. At the same time, this tends to create disturbances in the availability of these services for the local population – a perception which has been exacerbated during the COVID-19 shock (Kirişçi and Yavcan, 2019). A detailed analysis confirmed indeed that the arrival of refugees puts pressure in particular on the health system (numbers of doctors, midwives, hospitals and intensive care units per capita worsen). A 10 percentage-point increase in the refugee-to-native ratio, as is the case in several provinces, decreases the number of doctors per person by about 6–9 percent (Aygün et al., 2020). The gap between Syrian refugees’ and locals’ fertility rates (5.3% vs. 2.1%) appears to compound the perceived social challenges (Hacettepe University, 2019).

Another essential regional concern arises from geological risks. Earthquake threats are particularly high in the Istanbul region which hosts 19% of the population and generates 31% of GDP. The prevailing scientific view is that a quake of a magnitude 7-7½ on the Richter scale is statistically expected (Geomar, 2019; Swiss Re, 2015). A significant share (60% according to some estimations) of residential and non-residential buildings have been constructed semi-formally, without adequate and complete building authorisations. An urban transformation programme was launched after the 1999 earthquakes, which aimed at upgrading a large number of dwellings in the Istanbul area and in other risk-prone regions. The compliance of large parts of the building stock with earthquake resilience norms- which have themselves evolved through time- is nevertheless not assured. About 40% of Turkish households find that their dwelling would not resist to a serious seismic shock (Kadir Has University, 2020).

International reinsurers state that, despite good progress in insurance coverage after the 1999 earthquakes, only 40% of Turkey’s building stock is insured. They suggest that nearly 7 million people may be affected by a shock of magnitude 7.5, and material losses could attain USD 90 billion (Swiss Re, 2018). Natural disasters are traditionally covered by government self-insurance but OECD has adviced Member countries to offset financial risks by capital market instruments (OECD, 2017). Turkey created a National Catastrophe Insurance Pool (NCIP) which started to issue natural disaster bonds. The National Development Plan 2019-23 stated that additional earthquake risk mitigation strategies will be developed and implemented during the Plan period.

De-carbonising the economy and improving air quality

Reducing greenhouse gas emissions and local air pollution, and preserving non-renewable land and coastal resources are Turkey’s top environmental challenges (OECD, 2019g). Greenhouse gas emissions (GhG) are below OECD averages but grew at the fastest pace of the OECD area over the past decade (Figure 1.21). There was a degree of decoupling between emissions and economic activities, but, on current trends, emissions are expected to more than double between 2015 and 2030. Turkey has pledged, as an “intended national contribution” (INDC) commitment, to reduce GhG emissions by up to 21% from their business-as-usual level by 2030. While Turkey has signed, but not yet ratified, the Paris Agreement “as a developing country”, Turkish authorities continue to argue that the country’s developing country status should exempt it from net emission reduction targets – a request so far not acquiesced by international partners. The OECD Environmental Performance Review of Turkey in 2019 concluded that Turkey needs “a resilient development strategy that should integrate climate and energy objectives” and recommended the adoption of “a new National Climate Change Action Plan with sector-specific targets and monitoring mechanisms” (OECD, 2019g). The Ministry of Environment and Urbanisation started to work on such a plan in 2020 and aims at completing it in 2023.

Figure 1.21. The carbon intensity of the economy should be reduced

The share of renewable resources in the production of electricity reached 44% by the end of 2019, above the OECD average. This is mainly due to hydro-power sources. The underlying potential is significantly larger however, thanks to high solar exposition (Bankovic, 2019). The Ministry of Environment and Urbanisation is conducting studies on carbon pricing and on emission trading, in the context of the preparation of the new climate law, which could help better mobilise this potential.

Carbon pricing would help orient GHG containment efforts more efficiently. Fuel taxes in Turkey are among the highest in the OECD but are concentrated on certain types of road fuels. They serve fiscal rather than environmental objectives. The elasticity of road fuel demand to fuel taxes is very low (Erdogdu, 2013), and road-based emissions have continued to rise. Emissions other than from motor vehicles are taxed lightly (Figure 1.22). The gap between gasoline and diesel taxes (both are taxed heavily) encourages diesel use. Even if this gap narrowed in the recent period, diesel cars grew in popularity, contributing to high levels of air pollution (OECD, 2019g). At the same time, lighter taxation of liquefied gas (LPG) made it a popular alternative and has a positive impact on road pollution.

Figure 1.22. There is little or no carbon pricing in most sectors, except for road fuel

Average effective carbon rates across emissions by sectors, 2015 (Height of bar shows effective price of carbon. Width shows range of emissions covered.)

Managing a carbon credit system would facilitate transition to carbon pricing. Although Turkey is not yet implementing carbon pricing, it seeks carbon credits from international markets. The Ministry of Environment and Urbanisation explored a roadmap for an emissions trading system. Such a system would include a dynamic allowance reserve to allow for growth, would grandfather allowances with a certain share of auctioning, and would use domestic offsets registered under voluntary standards (Ecofys, 2016). The harmonisation of these early efforts with the EU Emissions Trading System (ETS) Directive has already permitted to establish a good emission monitoring, reporting and verification (MRV) infrastructure, covering half of total emissions.

Air quality is a major concern in Turkey, especially in large cities (EEA, 2018). Population exposure to particulate matter emitted by power generation, road transport and heating is well above the benchmarks recommended by the World Health Organisation (Figure 1.23). According to the European Environment Agency 2019 Report on Air Quality (EEA, 2019), annual mean concentrations of some of the pollutants (including PM.2.5 and NO2) are particularly high in Turkey. The Ministry of Environment and Urbanisation is running several studies and projects to improve air quality, but, for effective air quality management, policymaking should be supported by reliable information organised in formal models and according to benchmark indicators. The Ministry publishes air quality indicators according to EU norms on a dedicated website (Ministry of Environment and Urbanisation, 2020), this information should be further completed to cover all key sources of pollution in the entire territory.

OECD counselled to retrofit old coal power plants with state-of-the-art clean technology, or close them down. The government’s decision in 2019 not to delay the implementation of air pollution norms for coal power plants was a helpful step. Further, OECD recommends that coal should be gradually substituted with natural gas in residential heating, as currently envisaged by the authorities (OECD, 2019g). Policymakers also indicate that the thermal plants which do not meet pollution regulation benchmarks have not operated in the recent period. Limit values for air pollutants are planned to be aligned with EU standards by 2024. In transportation, cutting local air pollution calls for a modal shift from private to public transportation and cleaner road vehicles. The government is also sponsoring a national electrical car project. The infrastructure for electrical cars is being developed and the motor vehicles tax for electrical cars was reduced by 75%. The Ministry of Environment and Urbanisation aims also at building 3000 km of bicycle paths and 3000 km green of walking paths “to reduce air traffic pollution and to increase the physical and mental health of the population”.

Fine coastal areas are Turkey’s unique natural assets and are under pressure. Their depletion would hinder population’s well-being as well as the international competitiveness of the Turkish tourism industry. Pressures are particularly strong in the Marmara region around the Bosphorus. Land preservation policies are implemented only at a limited scale in Turkey (OECD, 2019g). Additional public information campaigns on the vulnerability of natural assets would improve national and local governments’ accountability for their protection.

Figure 1.23. Air quality and other local environmental conditions should improve

There is a large potential for further environment-enhancing investment in public infrastructures and for greening economic activities in the business sector. Mobilising this potential would support the recovery and help make growth more sustainable after the COVID-19 shock. OECD governments have already included “green recovery measures” in their post-COVID-19 policy packages and Turkey can follow suit (OECD, 2020i). While manufacturers benefit already from tax incentives for investing in energy-saving equipment, there is further potential for energy saving in buildings, which can be leveraged with tax support. Investment in this area is generally labour-intensive and new programmes can stimulate job creation.

The “European Green Deal Agenda” creates new opportunities and challenges for Turkey. This agenda aims at further decarbonising energy production and at promoting cleaner manufacturing and transportation in the EU and trade partners. It will likely imply border carbon taxes. Turkey is implementing a Partnership for Market Readiness (PMR) project, to prepare for these measures. A new legal framework is being developed for monitoring GhG emissions, in connection with the intended EU-compatible emission trading system. Progress in this area can help Turkey enlist EU support for transition investments - as is already partly the case (Corporate Sustainabilty and Responsibility Europe, 2020).

Upgrading public governance

OECD empirical research finds that effective governance institutions enhance growth and social inclusion beyond and above specific policy measures (Cournède et al., 2018). The National Development Plan 2019-23 stresses the need to reinforce the governance environment as the basis for other policies. Turkey’s position in international governance benchmarks suggests that there is still considerable room for progress in this area. Gaps vis-à-vis OECD and peer countries have increased. Shortcomings in the rule of law and in public sector integrity deserve special attention (Figure 1.24).

Figure 1.24. Room for progress in governance institutions exists

Percentile rank, 0 to 100 (higher the better)

Figure 1.25. Rule-of-law should be improved

Rule of law and selected sub-components, scale from 0 to 1 (strongest adherence to the rule of law), 2020

Improving on public integrity and fighting corruption and money laundering

Public integrity involves public trust in civil servants’ morality and capacity to resist private interests. It requires a low degree of high as well as petty corruption (OECD, 2018c). It has become critical in the COVID-19 context, as governments assume broader responsibilities, public finances come under pressure, and the depth of the pandemic is testing people’s trust in official institutions. Improving public integrity would also help attract new capital inflows, especially foreign direct investment in the context of the ongoing re-organisation of the global value chains (Gaspar et al., 2020).

Turkey has fallen behind OECD countries in various indicators of public integrity (Figure 1.26). The Integrity of the judiciary is a particularly pressing issue. In 2017, the OECD Working Group on Bribery (WGB) expressed its concern about Turkey’s lack of progress with respect to recommendations made in Phase 3 of its work in 2014, with only 3 out of 27 recommendations fully implemented (OECD, 2017b). The Phase 3 Report highlighted the WGB’s concern that foreign bribery investigations and prosecutions may be subject to improper influence by concerns of a political nature, and the Group decided to pay special attention to developments in this area, recommending that Turkey safeguards the independence of its judiciary and prosecution authorities (OECD, 2014) (see below regarding other key WGB recommendations not implemented with respect to transnational bribery). The WGB reiterated in a 2019 press release that it was “highly concerned that foreign bribery investigations and prosecutions may be influenced by considerations of national economic interest, the potential effect upon relations with another State or the identity of the natural or legal persons involved” (OECD, 2019).

In its 2019 Report on Turkey, the European Commission has noted “increased political interference that affects the quality and efficiency within the judicial system” (European Commission, 2019). While more judges were hired, the Commission identified shortcomings in objective and merit-based recruitment criteria. It found no legal guarantee to ensure the independence of the judiciary from political involvement. In its Compliance Report under the Fourth Evaluation Round, the Group of States against Corruption (GRECO), Council of Europe’s anti-corruption body, also raised concerns about lack of progress in this area since its previous evaluation, with the judiciary appearing to be less independent than before. The assessment found Turkey’s compliance with the recommendations to be ‘globally unsatisfactory’ (Council of Europe/GRECO, 2019). Turkish authorities indicated that, as a follow-up to these recommendations a Judicial Reform Strategy and a Turkish Declaration on Judicial Ethics were adopted. The Judicial Reform Strategy includes several measures such as geographical guarantees for judges and public prosecutors with a certain professional seniority, restructured promotion procedures for judges and public prosecutors based on qualifications and performance, and new rights for judges and public prosecutors during disciplinary processes.

In its latest plenary meeting on 26-29 October 2020, GRECO invited Turkish authorities to authorise the publication of its confidential reports in the area of incriminations and transparency of party funding as soon as possible. It found that in the area of corruption prevention Turkey’s compliance with its recommendations remained “globally unsatisfactory” (Council of Europe/GRECO, 2020).

There is no rigorous anti-corruption strategy currently in Turkey and administrative functions for integrity and anti-corruption remain ad hoc. This has allowed progress against corruption to stall (European Commission, 2019). A roadmap to strengthen the autonomy of independent bodies, to implement ethical training, and implement objective hiring criteria of public officials in the administration will be critical to improving public integrity.

Turkey has also taken limited steps to fight bribery of foreign public officials in international business transactions. As a member of the OECD Anti-Bribery Convention (OECD, 1997), Turkey is lagging behind on its commitment to enforce its foreign bribery laws, which has prompted the WGB to issue a press release in March 2019 (OECD, 2019e). Twenty years after the adoption of the Convention, the Working Group found that Turkey’s failure to observe key standards of the Convention undermines its ability to fight foreign bribery. In addition to issues around the independence of the judiciary and prosecution authorities, Turkey was urged to address deficiencies in its corporate liability regime, and to take measures to adequately protect whistleblowers. The Working Group also expressed its serious concern about Turkey’s low level of foreign bribery enforcement, noting the absence of any foreign bribery conviction since the entry into force of the Convention in Turkey.

Money laundering and terrorism financing should also be further strengthened given the threats in this area in Turkey’s region. Financial Action Task Force, in a 2019 report, invited Turkey to achieve “major improvements” in nine of eleven immediate outcomes (IOs) to counter these risks (FATF, 2019) (Figure 1.26, Panel F). It called Turkey to close the gaps in, notably, two areas: i) prioritisation of the use of financial intelligence on money laundering and developing a national strategy for investigating and prosecuting different types of money laundering, and ii) transposition of UN designations without delay. Progress is expected over a period of 16 months for Turkey to avoid being placed in a grey list (Reuters, 2019).

Figure 1.26. Risks of corruption, tax opacity and money laundering

Policy choices will be key to how Turkey emerges from the COVID-19 crisis and in the years ahead

The ability of Turkey to shift to a stronger and more inclusive growth path will depend on making policy and institutional reforms. Turkey’s own as well as international experience show that mainstream macroeconomic, microeconomic and institutional reforms generate very high returns, in particular in middle-income countries (Gönenç, 2017). An econometric analysis for this Survey finds that a package of structural reforms combined with stronger economic institutions could boost the level of GDP per capita by around 10% over 10 years as compared to a scenario with no policy changes. Table 1.4 sets out three reform scenarios, based on an OECD econometric model reflecting different degrees of implementation of the reform recommendations of this Survey:

The baseline trajectory is based on ongoing medium-term trends and assumes no changes to structural policies and economic institutions. It projects growth averaging 2.9% of GDP per year over the next 10 years, leading to an increase of the level of GDP per capita by around 33%. The estimation framework underpinning the baseline trajectory rests on data prior to the COVID-19 pandemic.
A first reform package would be centred on market liberalisation measures, including the removal of anticompetitive regulatory barriers in product markets (OECD, 2020m), more flexibility in labour markets, and cuts in labour and corporate income taxes. Over a 10 year horizon, the annual average growth rate of GDP per capita could increase by an additional 0.37% as compared to baseline with no policy changes.
A second reform package would prioritise institutional and educational reforms, including an upgrade in the quality of governance institutions and above-trend improvements in skills. Average annual GDP per capita growth could then increase by an additional 0.6%.
A third avenue would consist of an integrated package of market liberalisation and institutional and educational reforms (a combination of the two reform avenues above). Growth of GDP per capita would be nearly 1% above the baseline.

Scenario	Policy assumptions	Increase in average annual yearly growth of GDP per capita in 2015 PPP USD over 10 years
Scenario 1: Market reform package	Close the gap in product market regulations to the OECD average over 10 years. Close the gap in employment protection regulations to the OECD average of over 10 years. Decrease the labour income tax wedges to the average of the 5 best performing OECD countries over 5 years. Decrease corporate income tax rates to the level of the average of the 5 best performing OECD countries over 10 years.	0.37%
Scenario 2: Institutional and social progress package	Increase the quality of education. Increase the level of rule of law to the OECD average over 10 years.	0.54%
Scenario 3: Market reforms and institutional and social progress package	All reforms from scenario 1 and 2.	0.91%

Scenario

Policy assumptions

Increase in average annual yearly growth of GDP per capita in 2015 PPP USD over 10 years

Scenario 1: Market reform package

Close the gap in product market regulations to the OECD average over 10 years.

Close the gap in employment protection regulations to the OECD average of over 10 years.

Decrease the labour income tax wedges to the average of the 5 best performing OECD countries over 5 years.

Decrease corporate income tax rates to the level of the average of the 5 best performing OECD countries over 10 years.

0.37%

Scenario 2: Institutional and social progress package

Increase the quality of education.

Increase the level of rule of law to the OECD average over 10 years.

0.54%

Scenario 3: Market reforms and institutional and social progress package

All reforms from scenario 1 and 2.

0.91%

1. Model results need to be interpreted with caution. In some instances, the policy variables in the model captures the reform recommendations only very roughly hence estimates should be seen as illustrative. Policy changes in the model are based on comparing the policy settings in Turkey with other OECD countries. The model assumes that any spending increases are offset such that reforms are fiscally neutral. The model does not capture endogeneous policy-induced changes in deep-rooted preferences like risk aversion or time preferences..

Policy recommendations

Social and economic support against the COVID-19 shock

Continue to support workers and fundamentally sound firms in temporarily affected activities.
Replace the concessional loans and the one-off transfer to households at risk of poverty into a targeted allowance for a limited period.
Grant an across-the-board employer and employee social security contribution exemption to all young workers (15-24) for a temporary period.
Continue to strengthen vocational education.
Ensure that working-age recipients of unemployment benefits and other social supports actively look for jobs, and participate effectively to the re-training programmes on offer.

Macroeconomic policy for a sustainable recovery

Use the room available in public finances for transparent, temporary and targeted direct fiscal supports and resume fiscal tightening once the recovery is firmly underway.
Outline and communicate a coherent macroeconomic policy framework encompassing fiscal, quasi-fiscal, monetary and financial policies.
Publish a regular Fiscal Policy Report making transparent and projecting all public financial liabilities.
Restore the independence of the central bank, including with legislative measures.
Maintain the real policy interest rate in positive territory as long as inflation and inflation expectations diverge from official projections and targets.
Replenish foreign reserves as conditions allow.
Communicate actively on the foreign reserve position according to the information needs of financial markets.
Re-evaluate and reduce the weight of government-owned financial institutions.
Maintain a neutral framework for banks’ credit allocation decisions.
The authorities should communicate on how they evaluate and address the risks of deterioration in banks’ asset quality.
The results of the stress tests of individual banks and of the banking system as a whole should be disclosed to the public.
As long as the pandemic is not under control, use all available fiscal room for supporting the health system and the households and businesses in need. Lesser priority spending projects should be postponed, to preserve room for rapid fiscal response to changing needs
Discontinue tax amnesties and re-examine the recent digital taxes in the light of ongoing international co-operation in the OECD/G-20 process.
The Central Bank should spell out a strategy for rolling back, as soon as circumstances permit, the exceptional COVID-19 liquidity measures.
Resort to capital flow management measures only in exceptional circumstances and in accordance with the OECD Code of Liberalisation of Capital Movements.
Financial regulators should involve the Turkish Competition Authority to ensure a level playing field between public and private financial firms, as well as between public and private companies in access to finance.
Explicitly address various public concerns about statistical, methodology, data source and data quality issues related to all economic indicators, including those concerning inflation and monetary policy. Use central bank’s high-quality analysis and communication instruments such as the Inflation and Financial Stability Reports also for this purpose.

Improving employment and job quality

Continue to facilitate labour force participation of women, including by increasing the provision and quality of early child education.
Make fixed-term and temporary work contracts more flexible and the severance compensation system less costly.
Consider exempting employers from 50% to 100% of health insurance contributions and finance these cuts from other sources and savings.
Make statutory minimum wages more affordable for lower productivity firms, notably by promoting enterprise level bargaining for the settlement of minimum wage floors according to enterprise and regional conditions rather than one-size-fits-all national legislation.
Re-activate firm-level social dialogues, which are best suited to address COVID-19-related re-organisation challenges
Consider conducting an in-depth review of incentives to R&D to further boost R&D investment of businesses while ensuring a level-playing field for competition.

Institutional modernisation

Improve the quality of governance institutions and rule-of-law, with special focus on the independence and credibility of the judiciary, checks-and-balances over government powers, and a strategy of fight against corruption.
Promote public discussions on the perceived quality of governance institutions and degree of compliance with the rule-of-law on the basis of international benchmarks.
Comply with the recommendations of the Group of States against Corruption (GRECO).
Comply with the recommendations of the OECD Working Group on Bribery.
Comply with the recommendations of the 2019 Financial Action Task Force report on fighting money laundering and terrorism financing.

Environmental sustainability

Implement a carbon pricing policy, applicable gradually after the COVID-19 shock and encompassing all sectors.
Implement the recommendations of the 2019 OECD Environmental Performance Review of Turkey. In particular, adopt a new National Climate Change Action Plan as planned by the authorities.
Design a strategy to increase the share of renewable resources in the primary energy production, drawing notably on the solar potential.
Prepare and publish daily local air quality indicators according to international standards in the entire territory.
Develop a holistic strategy to improve air quality.
Consider tax reliefs for energy-saving investments in the building sector.
Continue to prepare the business sector to the introduction of border carbon taxes by trade partners.

Annex A. Turkey’s follow-ups to main past OECD recommendations

Past OECD recommendations	Follow-ups
Strengthening macroeconomic institutions
Publish quarterly general government accounts according to international national accounting standards and a regular Fiscal Policy Report covering all contingent liabilities and quasi-fiscal activities of the government.	The Ministry of Treasury and Finance, the Strategy and Budget Unit of the Presidency anf Turkstat publish general government accounts, on the basis of statistical data from the General Directorate of Accounting of the Ministry of Finance and Treasury, but according to slightly different methodologies which have their respective utilities. Turkstat started to publish them according to internationally comparable national accounting standards. Unifying the communication and monitoring of fiscal policy according to national accounting standards is recommended.
To reorient spending to top priority areas, implement the strategic and performance-oriented budgeting objectives of the Public Finance Law 5018.	The 2021 budget law was prepared as a program budget and was submitted to the Parliament on 17 October 2020.
Restore the credibility of monetary policy by committing all stakeholders to the independence of the central bank. Forward guidance should be provided on how the authorities plan to achieve the inflation target.	The Monetary Policy Committee pledges, in its press releases, to determine its future monetary stance on the basis of underlying inflation trends.
Streamline and stabilise business incentives. Report them according to state aid law, subject them to competition review, and monitor their impact on beneficiary firms’ behaviour using the new Enterprise Information System (EIS).	Research organisations started to conduct empirical analyses of the impact of business incentives on the basis of EIS information.
Streamline the various R&D incentives schemes on the basis of cost benefit analyses, and build on international best practices to improve take-up and efficiency of tax subsidies and grants.	Recent Research-Development-Innovation strategies started to draw on the economic impact analyses of earlier support programmes.
Product market reforms
Identify the remaining obstacles to the opening of network sectors to competition.
Delink agricultural support from production and shift its composition away from price measures towards direct support.
Labour market reforms
Liberalise fixed-term employment contracts.		A draft amendment to Labor Law was submitted to Parliament in October 2020 to make the fixed-term employment contracts for workers below 25 and above 50 more flexible, in order to increase their employment opportunities. It was withdrawn following strong union opposition. Consultations resumed with social partners.
Reform the severance payment system (to reduce its cost for employers and facilite the coverage and mobility of workers)		The 11th Development Plan 2019-2023 has set the severance pay reform as one of its main objectives.
Keep the growth of the official minimum wage below average productivity gains for a while.
Allow regional differentiation of minimum wages through local consultations between employers, employees and public sector representatives.		Regional minimum wages were implemented between 1951-1974 and the experience was not found to be positive. Following recent consultations between social partners a national minimum wage system is maintained.
Grant permanent social contribution cuts for low-skilled workers in the entire country, financing them by widening the tax base.		Social security contributon incentives have been applied to women and young workers since 2011 and to young entrepreneurs since 2018.
Facilitate further women’s labour force participation, notably by increasing the provision and quality of early child education and elderly care.		A legal amendment in July 2019 requested all firms employing more than 150 female workers to provide day care services to employees with children aged 0-6. A new “Women Up” project co-financed with the EU will support 4000 woman employers, with a budget of 30 million Euros. Other programmes aim at supporting the labour force participation of women with young children through institutional child care or professional caregivers.
Evaluate the uptake of the recent social security contribution cuts and make permanent those which have proven most supportive of formal sector job creation, financing this through better tax enforcement.		Existing temporary and permanent social security contribution cuts to foster job creation are being analysed by the Ministry of Treasury and Finance in coordination with the ministries in charge of implementation, in order to increase their effectiveness.
Education policy
Continue to reduce the quality gaps between schools.
Grant more autonomy to education institutions, against greater performance accountability.
Further develop pre-school education.				The project for “Increasing the Quality of and Access to Early Childhood Education”, co-financed by the EU and implemented with the technical support of UNICEF aims at updating early childhood education curricula and to facilitate access with the help of flexible participation models. The project will help to increase quality of and access to ECE services for 0-to-6-year-old children especially in the most vulnerable communities. It will be implemented in 20 provinces with a large population of 3-6 year old children with low participation to pre-school education.
Continue to strengthen vocational education in co-operation with enterprises and evaluate the outcomes of the recent initiatives in this area.				A vocational Training and Skills Development Cooperation Protocol (MEGİP) between the Turkish Employment Agency (İŞKUR) and the Union of Chambers and Commodity Exchanges of Turkey (TOBB) permitted to offer vocational training courses and on-the-job training programmes responding to business sector needs starting from 2018. As a result of this cooperation, there has been an increase in the number of trainees attending vocational training courses co-organised with the private sector.
Environmental protection and green growth
Use economic instruments such as harmonised pollution taxes and emission permits to reduce carbon emissions.			Turkey is not yet implementing carbon pricing but the Ministry of Environment and Urbanisation is conducting technical evaluations.
Evaluate and manage the environmental impact of massive transformations of land and sea.			1/100.000 scaled territorial plans in 19 planning regions (covering 61 provinces) have been approved by the Ministry of Environment and Urbanisation, designating residential and business investment areas on the basis of environmental protection objectives. Integrated Coastal Zone Plans covering about 82% of Turkey’s coasts were also prepared by the Ministry. Strategic Environmental Assessments (SEAs) are also enforced in the sectors covered by SEA regulations.
Public governance and integrity
Strengthen the rule of law, judiciary independence and step up the fight against corruption.

Annex B. Fiscal implications of the reform recommendations of this Survey

Policy recommendations with fiscal implications	Preliminary estimation of approximate fiscal costs
The one-off transfer of TRY 1000 to the 6 million households at risk of poverty can be converted into a temporary but targeted and recurrent allowance for 6 months.	Disbursement during 6 months would cost 0.8% of GDP.
The employment of young workers (15-24) can be made exempt of all employer and employee social security contributions (TRY 1280 per worker, reducing employer costs by 22% and increasing worker incomes by 19%) for a temporary period.	Assuming that 3.1 million young workers currently in employment and an additional 300 000 young workers for whom new jobs are created benefit, this measure would cost 0.9% of GDP per year of implementation. It is assumed that young workers are paid in average the official minimum wage.
Totality or part of employer health insurance contributions (7.5% of gross wages) can be funded from the general budget as a first step in labour market reforms to permanently reduce the labour tax wedge.	A total exemption of employer health insurance contributions would cost approximately 1.5% of GDP per year, at unchanged employment levels. A 50% exemption would cost 0.7% of GDP per year. Calculations are based on the average wage of registered workers in 2019.

Bibliography

Acemoğlu, D. and M. Üçer (2019),”High-Quality Versus Low-Quality Growth in Turkey: Causes and Consequences” CEPR Working Paper, No. DP14070.

Acemoğlu D. and M. Üçer (2020), “The Ups and Downs of Turkish Growth”, unpublished mimeo.

Adar, S. et al. (2020), Customs Union: Old Instrument, New Function in EU-Turkey Relations, SWP Comment, No.48.

Adrian, T. et al. (2020), “A Quantitative Model for the Integrated Policy Framework”, IMF Working Paper, No. 20/122.

Adrian, T. and G. Gopinath (2020), Toward an Integrated Policy Framework for Open Economies, IMF Blog.

Akat A.S and S.Gürsel eds. (2020), Turkish Economy at the Crossroads, World Scientific Publishing.

Akçiğit, U., Y.E. Akgündüz, S.M. Cilasun, S.M., E. Özcan-Tok, E. and F. Yılmaz (2020),”Facts on Business Dynamism in Turkey”, European Economic Review, Vol. 158.

Akçiğit, U. and Y. E. Akgündüz (2020), ‘’Turkiye´de COVID-19 Odakli Talep Şoklarinin Cografi ve Sektorel Yayilimi’’, Working Paper, Chicago University.

Akgündüz, Y.E., Aldan, A., Bagır, Y.K. and H. Torun (2019), “Job mobility in Turkey”, Central Bank Review, Vol. 19/3, pp. 83-91.

Akman, S. (2020), ‘’Ithalatta Ilave Gumruk Vergileri Ne Anlama Geliyor ?’’, TEPAV Degerlendirme Notu.

Alessi, L. and C. Detken (2018), ‘’Identifying excessive credit growth and leverage’’, Journal of Financial Stability, Vol. 35, pp. 215-225.

Algan, Y, (2018), ‘’Trust and Social Capital’’, in J.E. Stiglitz ed. (2018) For Good Measure, OECD Publishing, Paris.

Alper K., H. Kara and M. Yörükoğlu (2013), “Alternative tools to manage capital flow volatility”, B.I.S. Paper.

Andreasen, E., S. Bauducco and E. Dardati (2019), “Capital Controls and Firm Performance”, Working Paper Series, No. 331/2019, ILADES-UAH,

Ankara Tabip Odasi (2020), Pandemi Butun Agirligi ile Devam Ediyor, Ankara.

Aiyar, S., W. Bergthaler, J.M. Garrido, A. Ilyina, A. Jobst, K. Kang, D. Kovtun, Y. Liu, D. Monaghan, and M. Moretti (2015),”A Strategy for Resolving Europe’s Problem Loans”, IMF Staff Discussion Note.

Ari, A., S. Chen and L. Ratnovski (2020), ‘’COVID-19 and non-performing loans: lessons from past crises’’, Research Bulletin, No.71, European Central Bank.

Atabek Demirhan A. (2020), "Türk Imalat Sanayi Firma Performansi Analizi: Diş Tcaret Faaliyetleri Açisindan Bir Inceleme", in "Uluslararası Ticarette Güncel Konular", İstanbul Kültür Üniversitesi yayınları (forthcoming).

Atiyas, İ., O. Bakış and Y. Orhan (2017),”Firm Dynamics and Job Creation in Turkey – Some Preliminary Results”, Working Paper, Sabanci University.

Avrupa Birliği Başkanlığı (2020), İşçi/İşveren temsilcilerinin bilgilenme, istişare ve katılımcılık sorumluluklarının güçlendirilmesi teklif çağrısı, Ankara.

Aygün, A., M.G. Kirdar and B. Tuncay (2020), ‘’The Effect of Hosting 3.4 Million Refugees on the Health System in Turkey and Infant, Child, and Elderly Mortality Among Natives’’ Koc University-TUSIAD Economic Research Forum Working Paper Series, No. 2014.

Bagir, Y. K. (2018), ‘’Impact of the Syrian Refugee Influx on Turkish Native Workers’’, Central Bank Review, Vol. 18/4, pp. 129-147.

Bankovic, A. (2019), Potential for the Future: Renewable Energy Investments in Turkey, EEL Events Blog.

BBVA Research (2020), Turkey GDP is slumped but beat market expectations, 31 August, (Link).

BBC News (2020a), Yuz yuze egitim; Egitimciler kademeli olarak ilerleyecek surec icin ne diyor?, (Link).

BBC News (2020b), Tukenen saglik calisanlari topluca istifa ediyor, (Link).

Benigno et al. (2020), Credible emerging market central banks could embrace quantitative easing to fight COVID-19, CEPR.

Bircan, C. and O. Saka (2019), ‘’Lending cycles and real outcomes: costs of political misalignment’’, Working Paper, No. 225, EBRD.

Bloom, N. and J. van Reenen (2010),”Measuring and Explaining Management Practices across Firms and Countries”, Journal of Economic Perspectives, Vol. 24 (1).

Boeri, T. et al. (2019), “Wage Equalization and Regional Misallocation: Evidence from Italian and German Provinces”, Discussion Paper Series, No. 12279, IZA Institute of Labor Economics.

BRSA (2020a), Press Release (About Asset Ratio), Banking Regulation And Supervision Agency, (Link)

BRSA (2020b), Press Release (About Derivative Transactions), Banking Regulation And Supervision Agency, (Link)

BIS (2017), Supervisory and bank stress testing; range of practices, Banking Regulation And Supervision Agency, (Link).

Çakmakli C. and S. Demiralp (2020), “A Dynamic Evaluation of Central Bank Credibility”, Koç University-TUSIAD Economic Research Forum Working Papers.

CBRT (2019a), Monetary and Exchange Rate Policy for 2020, Central Bank of the Republic of Turkey.

CBRT (2019b), Financial Stability Report November 2019, Central Bank of the Republic of Turkey.

CBRT (2020b), Financial Stability Report November 2020, Central Bank of the Republic of Turkey.

CBRT (2020a), House Price Index: September 2020, Central Bank of the Republic of Turkey.

Ceritoglu, E. et al. (2017), ‘The Impact of Syrian Refugees on Natives' Labor Market Outcomes in Turkey: Evidence from a Quasi-Experimental Design’, CBRT Working Papers, No. 17/05.

Cilasun S., R. Gönenç, U. Özmen, Z. Samancıoǧlu, F. Yilmaz and V. Ziemann (2019), “Upgrading business investment in Turkey”, OECD Economics Department Working Papers, No. 1532.

Citibank (2020), CEEMEA Economics: Turkey Economics, 17 March.

Cherif, R. and F. Hasanov (2019),”The Return That Shall Not Be Named: Principles of Industrial Policy”, IMF Working Paper, WP/19/74.

Council of Europe/GRECO (2018), Fourth Evaluation Round: Corruption prevention in respect of members of parliament, judges and prosecutors, Compliance Report Turkey, (Link).

Council of Europe/GRECO (2020), Plenary Meetings, (Link)

Cournède, B., J. Fournier and P. Hoeller (2018), ‘’Public finance structure and inclusive growth’’, Economic Policy Paper, No. 25, OECD Publishing, Paris.

Corporate Sustainability and Responsibility Europe (2020), The European Green Deal and Turkey, (Link).

Çetinkaya, Ö. (2014). “Kamu İhale Kanunu ve Yolsuzluk Riskleri”. Bursa: Ekin Yayınevi.

Çetinkaya, Ö., E. Eroğlu (2020),”An Assessment of the Public Procurement System in Turkey”, Public Financial Management Reforms in Turkey: Progress and Challenges, Vol. 1.

Çolak et al. (2019), ‘’Monitoring and Forecasting Cyclical Dynamics in Bank Credits: Evidence from Turkish Banking Sector’’, Working Paper, No. 19/29, CBRT.

Del Carpio, X., and T. Taskin (2019),”Quality of Management of Firms in Turkey”, Jobs Working Paper No. 27, World Bank.

Demiralp S. (2020), “The Economic Impact of the COVID-19 on Turkey”, Middle East Institutte, (Link).

Demiralp S., H. Kara and P. Özlü (2012), “Monetary policy communication in Turkey”, European Journal of Political Economy, December.

Deutsche Welle (2020a), Pandemi Doneminde Turkiye´de Gocmen ve Multeciler, (Link).

Dünya (2018), Turizmin digital gelecegi Antalya´da konusuldu, (Link)

Dusundere, A.T. and I.A. Koyuncu (2020), ‘’Yeni Ilave Gumruk Vergileri Turkiye icin Ne Ifade Ediyor?’’

TEPAV, Degerlendirme Notu.

EBA (2020), Milli Egitim Bakanligi Egitim Bilisim Agi, (Link).

Ecofys (2016), “Roadmap for the Consideration of Establishment and Operation of a Greenhouse Gas Emissions Trading System in Turkey”, Report for the Ministry of Environment and Urbanisation, (Link).

Eğilmez, M. (2020a), Merkez Bankası Rezervlerinin Yeterliliği, Blog.

Eğilmez, M. (2020b), Genc Issizlik Orani, Blog.

Estevão, M. and G. Everaert (2016), Enhancing Monetary Policy Flexibility Through `De-dollarization’, IMF Blog.

EpiForecasts (2020), Covid-19 Reproduction Number Estimates for Turkey, (Link).

Ergun, D., O. Gouveia and A. Ortiz (2017), International experiences of De-dollarization: What could be done in the Turkish case?, BBVA Research.

Erdogdu, E. (2013), ‘’Motor fuel prices in Turkey’’, MPRA Paper, No. 55521.

Egitimsen (2020), Salgin Gunlerinde Uzaktan Egitim Çalistayi Sonuc Raporu, 4 Eylul 2020.

Ernst & Young (2020a), Turizm Sektoru Degerlendirmesi 2020 Raporu, (Link).

Ernst & Young (2020b), Tourism Market Overview, Turkey & Istanbul, (Link).

Euronews (2020b), Turkiye´de kayit disi calisirken issiz kalan multeciler icin yardim cagrisi, (Link).

European Bank For Reconstruction and Development (2020), COVID-19; From Shock to Recovery, Regional Economic Prospects, (Link)

European Commission (2017), “Effectiveness of tax incentives for venture capital and business angels to foster the investment of SMEs and start-ups”, Final Report TAXUD/2015/DE/330.

European Banking Authority (2020), EU-wide stress testing, (Link).

European Commission (2019), Key findings of the 2019 Report on Turkey, (Link).

European Commission (2020), Key findings of the 2020 Report on Turkey, (Link).

EEA (2018), European Environment Agency Report on Air Quality in Europe 2018, (Link)

EEA (2019), European Environment Agency Turkey Air Pollution Country Fact Sheet 2019, (Link)

Faccio, M. (2006),”Politically Connected Firms”, The American Economic Review, Vol. 96 (1).

FATF (2019), “Anti-money laundering and counter-terrorist financing measures in Turkey”, Financial Action Task Force, (Link).

Gevrek, Z.E., P. Kunt and H.W. Ursprung (2019), ‘’Education, Political Discontent, and Emigration Intentions; Evidence from a Natural Experiment in Turkey’’, CEsifo Working Papers, No. 7710.

Gürakar, E. Ç. (2016), Politics of Favoritism in Public Procurement in Turkey: Reconfigurations of Dependency Networks in the AKP Era, Palgrave Macmillan.

Gaspar, V., M. Muhleisen and R. Weeks-Brown (2020), Corruption and COVID-19, IMF Blog.

Gaspar, V. and G. Gopinath (2020), Fiscal Policies for a Transformed World, IMF Blog.

Geomar (2019), Istanbul: Seafloor study proves earthquake risk for the first time, (Link).

Goldman Sachs (2020), CEEMEA in Focus: Turkey, 7 February.

Gönenç, R., S. Şahinoz and Ö. Tuncel (2010), ‘’Turkey´s Improving Integration with the Global Capital Market: Impacts on Risk Premia and Capital Costs’’, OECD Economics Department Working Papers, No. 812.

Gönenç, R. (2017), “The Middle-Income Plateau: Trap or Springboard?, OECD Economics Department Working Papers, No. 1446.

Gül, S. (2019). “Institutional Quality and the Sovereign Risk Premia: Cross Country Evidence”, Unpublished mimeo, Ankara.

Gül, S. (2020a). “Sovereign Risk and Local Currency Lending Rates: Evidence from Five OECD Countries”, CBRT Research Notes in Economics (forthcoming), Central Bank of the Republic of Turkey.

Gül, S. (2020b). “Determinants of Debt Dollarization for NFCs: Does the Institutional Quality Matter?”, CBRT Research Notes in Economics (forthcoming), Central Bank of the Republic of Turkey.

Gürsel S. and G. Uysal (2020), «Korona Gunlerinde Turkiye Isgucu Piyasasi », in O.F.Çolak ed. (2020) « Korona Krizinde Turk Ekonomisi », Elif Yayinlari (forthcoming).

Hacettepe University (2019), “Suriyeli Göçmen Örneklemi Temel Bulgular”, Nufus Etutleri Enstitusu, (Link)

Huizinga, H. and L. Laeven (2019), ‘’The Procyclicality of Banking: Evidence from the Euro Area’’, Discussion Paper Series, No. 13605, CEPR.

International Labour Organisation (2020), COVID-19 crisis and the informal economy: Immediate responses and policy challenges, (Link).

IMF (2019), 2019 Article IV Consultation with Turkey.

IMF (2020), 2020 External Sector Report: Global Imbalances and the COVID-19 Crisis.

Irwin, D.A. (2020), ‘’The Rise and Fall of Import Substitution’’, Working Paper, No. 27919, NBER.

Istanbul Buyuksehir Belediyesi (2020), Internet Erisiminde En Buyuk Sorun Pahalilik, (Link).

Kadir Has University (2020), Turkiye Egilimleri-2019 (M. Aydin et al.), (Link).

Kara H., T. Tiryaki and C. Yüksel (2014), “In Search of a Reasonable Credit Growth Rate for Turkey,” Central Bank Review 14 (1).

Kirisçi, K., and B. Yavcan (2020), “As COVID-19 worsens precarity for refugees, Turkey and the EU must work together”, Brookings, (Link)

KPMG (2020), Perakende Sektorel Bakis, (Link).

Mehr News Agency (2020), Turizm gelirlerinin üçte birine düşmesi büyümeyi 4 puan indirebilir, (Link).

Ministry of Development (2018), Saglik Hizmet Kalitesi ve Mali Surdurulebilirlik, 11th Development Plan (2019-2023), Report of the Special Commission, (Link).

Ministry of Environment and Urbanisation (2020), Ulusal Hava Kalite Izleme Ağı, (Link).

Ministry of Family, Labor and Social Services of Turkey (2020a), Isyerlerinde COVID-19 ile Mucadele, (Link).

Ministry of Treasury and Finance (2020), New Economy Programme 2021-2023.

Ministry of Treasury and Finance (2020a), Debt Indicators.

Mimir, Y. and E. Sunel (2019), ‘’External Shocks, Banks, and Optimal Monetary Policy: A Recipe for Emerging Market Central Banks’’, International Journal of Central Banking, Vol. 15/2, pp. 235-299,

National Geographic (2020), “Turkey has been contact tracing for a century. That offers lessons and perils”, (Link)

OECD (1997), Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, OECD Publishing, Paris.

OECD (2008), OECD Guidance on Sovereign Wealth Funds, (Link).

OECD (2010), OECD Economic Surveys: Turkey 2010, OECD Publishing, Paris.

OECD (2012), Recommendation of the Council on Regulatory Policy and Governance, OECD Publishing, Paris.

OECD (2013), Supporting Investment in Knowledge Capital, Growth and Innovation, OECD Publishing, Paris.

OECD (2014), Phase 3 Report on Implementing the OECD Anti-Bribery Convention in Turkey, OECD Publishing, Paris.

OECD (2015), Taxation of SMEs in OECD and G20 Countries, OECD Tax Policy Studies, No. 23, OECD Publishing, Paris.

OECD (2015b), Opportunities and Constraints of Market-Based Financing for SMEs, OECD Report to G20 Finance Ministers and Central Bank Governors.

OECD (2015c), OECD Guidelines on Corporate Governance of State-Owned Enterprises, OECD Publishing, Paris.

OECD (2016), OECD Economic Surveys: Turkey 2016, OECD Publishing, Paris.

OECD (2017),”Enhancing The Contributions of SMEs in a Global and Digitalised Economy”, Meeting of the OECD Council at Ministerial Level, Paris 2017.

OECD (2017), OECD Recommendation on Disaster Risk Financing Strategies, OECD Publishing, Paris.

OECD (2017b), Turkey Follow Up-To The Phase 3 Report & Recommendations, OECD Publishing, Paris.

OECD (2018), Latin American and Caribbean Competition Forum: The Informal Economy in Latin America and the Caribbean: Implications for Competition Policy, Background Note.

OECD (2018a), OECD Economic Surveys: Turkey 2018, OECD Publishing, Paris.

OECD (2018b), Good Jobs for All in a Changing World of Work, OECD Publishing, Paris.

OECD (2018c), OECD Strategic Approach to Combating Corruption and Promoting Integrity, OECD Publishing, Paris.

OECD (2019a), PISA 2018 Country Note: Turkey (Link).

OECD (2019b), “R&D Tax Incentives: Turkey, 2019”, Directorate for Science, Technology and Innovation, December, (Link).

OECD (2019c), OECD Pensions Outlook, OECD Publishing, Paris.

OECD (2019d), OECD leading multilateral efforts to address tax challenges from digitalisation of the economy, (Link). OECD (2019d), Pensions at a Glance 2019: OECD and G20 Indicators, OECD Publishing, Paris.

OECD (2019e), “Turkey’s foreign bribery enforcement framework needs to be urgently strengthened and corporate liability legislation reformed”, (Link).

OECD (2019f), OECD Family Database: Child poverty, (Link).

OECD (2019g), OECD Environmental Performance Reviews: Turkey 2019, OECD Environmental Performance Reviews, OECD Publishing, Paris. OECD (2019h), OECD leading multilateral efforts to address tax challenges from digitalisation of the economy, (Link).

OECD (2020), OECD Employment Outlook 2020: Worker Security and the COVID-19 Crisis, OECD Publishing, Paris.

OECD (2020a), “Corporate sector vulnerabilities during the Covid-19 outbreak: assessment and policy responses”, Tackling Coronavirus Series.

OECD (2020b), “Insolvency and debt overhang following the COVID-19 outbreak: assessment of risks and policy responses”, Tackling Coronavirus Series (forthcoming).

OECD (2020c), Coronavirus Country Policy Tracker: Turkey, (Link).

OECD (2020d), “Testing for COVID-19: A way to lift confinement restrictions”, Policy Responses To Coronavirus, (Link)

OECD (2020e), Capacity for Remote Working Can Affect Lockdown Costs Differently Across Places, Policy Responses To Coronavirus (Link).

OECD (2020f), OECD Code of Liberalisation of Capital Movements, (Link).

OECD (2020g), Job retention schemes during the COVID-19 lockdown and beyond, Tackling Coronavirus Series.

OECD (2020h), Capacity for remote working can affect lockdown costs differently across places, OECD Policy Responses to Coronavirus, (Link).

OECD (2020i), The OECD Jobs Strategy: Implementation Highlights for Korea.

OECD (2020j), How´s Life? 2020, OECD Publishing, Paris.

OECD (2020k), OECD Family Database: Intimate Partner Violence Indicators, (Link).

OECD (2020l), Making the green recovery work for jobs, income and growth, Policy Responses to Coronavirus.

OECD (2020m), Competition Assessment Toolkit, (Link).

OECD (2020n), OECD Economic Surveys: Costa Rica 2020, OECD Publishing, Paris.

OECD/ETF/EU/EBRD (2019), “SME Policy Index: Western Balkans and Turkey 2019: Assessing the Implementation of the Small Business Act for Europe”, SME Policy Index, OECD Publishing, Paris.

Oyak Yatirim (2020), Gelecek 12 ayda beklenen kârlara göre hesaplanan fiyat kazanç oranları, (Link).

Özlale, Ü. and B. Polat (2019),”Understanding Firm Dynamics and Job Creation In Turkey Using the Entrepreneur Information System Database”, TÜSİAD – Sabancı Üniversitesi Rekabet Forumu Working Paper No: 2019-2,.

Özatay, F. (2019), ‘’Mind the Gap: Debt Limit and Fiscal Stimulus Debt Limit’’, TEPAV Working Paper, No. 201901.

Özmen U. (2019) “Economic Complexity and Sovereign Risk Premia”, Economics Bulletin, Vol (39) 3.

P.A. Turkey (2020), Shoping mall stores debating closure, (Link)

Presidency of Strategy and Budget (2019), Eleventh Development Plan 2019-2023, Ankara.

Reuters (2019), Turkey must step up measures against money laundering, watchdog says,

Reuters (2020a), Turkey has only been publishing symptomatic coronavirus cases, (Link)

Reuters (2020b), Turkey's coronavirus cases total 1.749 million since outbreak began, (Link)

Sameks (2020), Satinalma Mudurleri Endeksi, Kasim 2020, MUSIAD (Link)

Sayan S. (2020), “Cari açık ilelebet payidar kalacak mı?”, TEPAV Arastirma Notu.

Sak G. and F. Özatay (2020), “What If Turkey Imposes a Curfew Due To COVID-19 ? », TEPAV Arastirma Notu.

SWF (2008), International Forum of Sovereign Wealth Funds Santiago Principles, (Link).

Swiss Re (2015), Risky Cities; Istanbul, (Link).

Swiss Re (2018), Risky Cities; earthquake resilience in Istanbul, (Link).

Tükel H. (2020), “COVİD 19 ve Türkiye’de İş Piyasası”, SEDEFED Uzman Görüşü Notu, October (Link).

Trading Economics (2020), Turkey Households Debt to GDP Ratio, (Link).

Tümen, S. (2019), ‘’The Effect of Refugees on Native Adolescents´ Test Scores; Quasi-Experimental Evidence from PISA’’, Economic Research Forum Working Paper, No. 1356.

Turkstat (2020), Uluslararasi Göç Istatistikleri: 2019, (Link).

Turkish Medical Association (2020a), ‘’Saglik calisanlarinin COVID-19 virusune maruz kalimina iliskin risk degerlendirmesi anketinin on raporu’’, (Link).

Turkish Medical Association (2020b), Hekimlerin Istifasi Yasaklandi mi? (Link).

Türk-Iş (2019), Suriyeliler ve Turkiye Isgucu Piyasasina Etkileri, (Link).

Türkonfed (2020), Dijital Liderler ve KOBI’ler donusum seminerleri, (Link).

TVF (2019), Turkey Wealth Fund and its subsidiaries, consolidated statement of financial position at 31 December 2019, (Link).

TVF (2020), The Mandate of the Turkey Wealth Fund (Link).

United Nations (2019), International migrant stock 2019: Country Profiles, (Link).

Webrazzi (2020a), Covid-19 salgini e-ticaret ve altyapi talebini nasil degistirdi? (Link).

Webrazzi (2020b), Türkiye'de e-ticaret yüzde 39 büyüyerek 83,1 milyar lira oldu, (Link).

World Bank (2014),”Evaluation of the EU - Turkey Customs Union”, (Link).

World Bank (2020),”COVID-19 Outbreak: Implications on Corporate and Individual Insolvency”, COVID-19 Notes Finance Series, World Bank Group.

WTO (2019), Global Value Chain Development Report 2019: Technological Innovation, Supply Chain Trade and Workers in a Globalised World, Lausanne.

World Justice Project (2019), Rule of Law Index 2019. (Link).

Yarba, İ., Z. N. Güner (2019),”Leverage dynamics: Do financial development and government leverage matter? Evidence from a major developing country”, Empirical Economics, May 2019.

Yilmaz, H.H. and S. Tosun (2010), 5018 Sayılı Kanun Çerçevesinde Mali Saydamlık ve Parlamentonun Bütçe Sürecinde Etkinliği, Yasama Dergisi 14.

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Countries

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Publications

Publications

Featured publications

Data

Data

Featured data

News & Events

News & Events

Featured Events

About OECD

About

Engage with us

Work with us

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment