25 January 2017 Turkish Equities Strategy

Hoist your sail when the wind is fair y

 Economy activity to normalise with expansionary fiscal policy: Our 12-month forward looking Index target underlying assumptions for 2017 consist of some stabilisation in the political BIST-100 actual level Consensus index target scene after the expected constitutional referendum in the first half of 2017, UNLU & Co index target 100,000 96,350 stabilisation in TL following the interest rate hikes by CBRT, and geopolitical 95,000

risks declining moderately. We forecast 3.0% GDP growth in 2017E, driven 90,000 91,080 mainly by higher government spending and a recovery in private 85,000 | CEEMEA/Turke consumption. On the other hand, inflation is likely to remain elevated 80,000

throughout the year. The main downside risks are i) a sustained rally in 75,000 Strategy insight energy prices, ii) geopolitical risks with terrorism hitting touristic areas, iii) 70,000 the continuing purge of the failed coup-plotters, iv) expected public Jul-15 Jul-16 Jul-17 Jan-15 Jan-16 Jan-17 Jan-18 Apr-15 Apr-16 Apr-17 Oct-15 Oct-16 referendum in 1H17 and v) Fitch lowering Turkey’s sovereign credit rating Oct-17 one notch to non-investment grade. Top picks for 2017 Ticker Rating CP (TL) TP (TL) Upside  Turkey is attractively valued in the EM context: If we witness further AKBNK Buy 7.84 9.40 20% inflows to emerging markets, which started in July 2016, Turkey could GARAN Buy 7.81 9.75 25%

potentially be among the beneficiaries as it trades at a significant discount to CCOLA Buy 37.56 43.90 17% Equity Research its peers. Turkey’s discount to the MSCI EM Index widened to an all-time ENKAI Buy 5.82 6.92 19% high in 2016 on the back of rising political risks and the weakening economic TUPRS Buy 78.00 91.00 17% outlook. Earnings, on the other hand, remained relatively resilient thanks to TTRAK Buy 78.95 92.00 17% the banking sector’s record high profitability in 2016, which sent Turkish ULKER Buy 16.61 22.30 34% equities’ multiples to multi-year lows. In 2017, we expect a recovery in Source: Bloomberg, UNLU & Co estimates Turkish non-financials earnings, while banks are likely to grow earnings further (see page 25). We believe the depressed level of multiples imply that investor appetite towards Turkish assets is at an all-time low. Research Analysts Vedat Mizrahi, PhD  Our bottom-up index target of 96,350 offers 16% upside: [email protected] Considering global trends and the macro outlook in Turkey, we expect TL +90-212-367-3690 interest rates to remain elevated. As a result, we adjust our risk-free rate Can Ozguzel assumption to 10.5%, from 9.5% previously. In this report, we revise our [email protected] target prices and recommendations for all the companies in our coverage +90-212-367-3678 based on the changes to our macro forecasts, COE assumptions and

earnings estimates. Our bottom-up index target is 96,350, offering 16% Can Oztoprak upside potential. We view relatively more attractive upside potentials in the [email protected] banks, real estate and conglomerates sectors. On the other hand, the retail, +90-212-367-3692 TMT, automotive, white goods, cement, aviation, glass and defence sectors Emir Moran offer relatively less attractive upsides potentials (See Figure 46 on page 23, for [email protected] all the updated rating and TPs). +90-212-367-3687

 Top picks: Based on our macro outlook for 2017, we identify five main Esra Simsek investment themes that could play out well in 2017. We discuss each of [email protected] these themes in detail and determine 12 preferred names around the +90-212-367-3696 themes (see page 27). To summarise; i) banks outshine non-financials from a Mete Ozbek relative risk/reward perspective, ii) hedge currency risk through exporters [email protected] and FX earners, iii) high dividend payers should outperform, at least during +90-212-367-3689 the dividend season, iv) seeking recovery and restructuring stories and v)

generating alpha through quality small-caps Among these 12 ideas, we name Muharrem Gulsever the most liquid ones as our Top picks. Accordingly, Akbank, Garanti, [email protected] Coca-Cola Icecek, Enka Insaat, Tupras, Turk Traktor and Ulker are our +90-212-367-3694 most preferred names (see pages 35-43). Sinan Veziroglu [email protected] +90-212-367-3691

IMPORTANT DISCLOSURES AND ANALYST CERTIFICATIONS ARE IN THE DISCLOSURE APPENDIX. U.S. Disclosure: UNLU Menkul Degerler A.S. ("UNLU & Co") does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the Firm may have a conflict of interest that could affect the objectivity of this report. Investors should not consider this report as the only factor in making their investment decision. U.S. investors transacting in the securities featured or mentioned in this research report must deal directly through a U.S. Registered broker- dealer.

Contents Executive summary ...... 3 2016 was a year of two distinct halves ...... 3 US policies and the performance of USD will be key determinants ...... 3 Economy activity to normalise with expansionary fiscal policy ...... 3 Turkey is attractively valued in the EM context ...... 3 Our bottom-up index target (96,350) offers 16% upside ...... 4 Top picks for 2017 ...... 4 Macroeconomic outlook ...... 6 Weak growth in 1H17, followed by recovery ...... 6 Unemployment rate rises significantly ...... 7 Inflation is likely to surge in 1H17 ...... 8 CBRT prefers unorthodox measures to tighten ...... 9 Current account deficit to widen in 2017 ...... 10 Budget deterioration in sight, though not an immediate concern ...... 11 Private sector is highly leveraged ...... 12 Political risks weigh on the economy ...... 13 Fixed income market outlook ...... 14 Turkey could lose its last IG rating from Fitch Ratings ...... 14 CDS increase sharply after the attempted coup ...... 15 Sovereign USD bond yields increase by c.80bps in 2016...... 15 Equity market outlook ...... 16 Current valuations reflect a crisis scenario ...... 16 De-rating has primarily been driven by banks ...... 18 The value discount of non-banks is also on the rise ...... 19 Key revisions and rating changes ...... 20 We revisit our entire coverage universe’s valuation ...... 20 Revised COE leads to an 11% average cut in our TPs ...... 21 Our bottom-up BIST 100 index target is 96,350 ...... 22 We change the ratings of 18 companies ...... 22 Consensus revision trends ...... 24 Earnings revisions favour the banks ...... 24 Looking for 25% EPS growth for TR equities ...... 25 EBITDA growth accelerates in 2017 with normalisation in some sectors ...... 25 Earnings revisions: How does Turkey stand in the EM context? ...... 26 Key themes for 2017 ...... 27 Theme 1: Banks outshine non-financials from a relative risk/reward perspective ...... 28 Theme 2: Hedge currency risk through exporters and FX earners ...... 31 Theme 3: High dividend payers should outperform, at least during the dividend season ...... 32 Theme 4: Seeking recovery and restructuring stories ...... 34 Theme 5: Generating alpha through quality small caps ...... 34 Top picks ...... 35 Reviewing the performance of 2016 calls ...... 35 UNLU & Co Top Picks for 2017 ...... 36 Akbank: BUY; 12M target price, TL9.40/sh ...... 37 Garanti: BUY; 12M target price, TL9.75/sh ...... 38 Coca-Cola Icecek: BUY; 12M target price, TL43.90/sh ...... 39 Enka Insaat: BUY; 12M target price, TL6.92/sh ...... 40 Tupras: BUY; 12M target price, TL91.00/sh ...... 41 Turk Traktor: BUY; 12M target price, TL92.00/sh ...... 42 Ulker Biskuvi: BUY; 12M target price, TL22.30/sh ...... 43

Equity Research 2

Executive summary 2016 was a year of two distinct halves

2016 was a year of two distinct halves for Turkey. In line with our expectations, strong economic activity in 1H16 transitioned to a considerable slowdown in the second half of the year amid political uncertainties, which was further exacerbated by the failed coup attempt in mid-July. Despite the Central Bank’s monetary easing to boost loan demand, economic activity failed to pick up through the rest of the year amid the declining real estate sector and consumer confidence on the back of a sharp devaluation in the Turkish Lira. Rising inflation, coupled with loose monetary policy, left the market with almost zero ex-post real interest rates in Turkey in an environment where the global markets were convinced that the US Federal Reserve (Fed) would raise policy rates on the back of their strengthening economy. This, along with political uncertainties, created a confidence crisis and led to a massive currency devaluation. US policies and the performance of USD will be key determinants

With US economic growth picking up from its subdued pace, we assume three rate hikes from the Fed in 25bps instalments, in line with the market consensus. While this expectation has resulted in USD strength since the US elections, we have witnessed some normalisation lately, which triggered investors to tiptoe back into emerging market stocks. Last year, the benchmark MSCI EM Index snapped a three- year losing streak, but ended 2016 with its worst quarter in more than a year as the USD Index climbed to its highest level in more than a decade. It seems that US President Trump’s policies and the course of the US dollar are likely to be the most important determinants of the performance of emerging markets. Economy activity to normalise with expansionary fiscal policy

On the backdrop of local and global risks, our underlying assumptions for 2017 consist of some stabilisation in the political scene after the expected constitutional referendum in the first half of 2017, stabilisation in TL following the interest rate hikes by CBRT, and geopolitical risks declining moderately. We forecast 2.4% GDP growth in 2016E, followed by 3.0% in 2017E, driven mainly by higher government spending and a recovery in private consumption. Inflation is likely to remain elevated throughout the year and end the year flat at 8.5%. The current account deficit will once again widen and its financing will remain critical. The government’s growth- enhancing fiscal policies will result in a deterioration in the budget deficit, but it will not become an area of major concern for now (see pages 6-13).

The current downside risks to the Turkish economy include a sustained rally in energy prices, geopolitical risks with terrorism hitting touristic areas, the continuing purge of the failed coup-plotters, expected public referendum in 1H17 and Fitch lowering Turkey’s sovereign credit rating one notch to non-investment grade. Turkey is attractively valued in the EM context

If we witness sustained inflows to emerging markets, Turkey could potentially be among the beneficiaries as it trades at a significant discount to its counterparts. Turkey’s discount to the MSCI EM Index widened to an all-time high in 2016 on the back of rising political risks and the weakening economic outlook (see page 16). Earnings, on the other hand, remained relatively resilient thanks to the banking sector’s record high profitability in 2016, which sent Turkish equities’ multiples to multi-year lows. In 2017, we expect a recovery in Turkish non-financials earnings, while banks are likely to grow earnings further (see page 25). We believe the

Equity Research 3

depressed level of multiples imply that investor appetite towards Turkish assets is at an all-time low. Our bottom-up index target (96,350) offers 16% upside

Considering global trends and the macro outlook in Turkey, we expect TL interest rates to remain elevated. As a result, we adjust our risk-free rate assumption to 10.5%, from 9.5% previously. While a 100bps increase in our risk-free rate assumption results in an 11% decline in target values for the companies under our coverage, an upgrade to our earnings estimates (especially for the banking stocks) offsets this negative impact. In this report, we revise our target prices and recommendations for all the companies in our coverage based on the changes to our macro assumptions, COE assumptions and earnings estimates. Accordingly, our bottom-up index target for the next 12 months is 96,350, which offers 16% upside potential. While this may not look attractive, we view relatively more attractive upside potentials in the banks, real estate and conglomerates sectors. On the other hand, the retail, TMT, automotive, white goods, cement, aviation, glass and defence sectors offer relatively less attractive upside potentials.

We upgrade five companies from Hold to Buy and two companies from Sell to Hold, on the other hand, we downgrade 10 companies from Buy to Hold and one company from Hold to Sell due to relatively limited upside potentials. In Figure 1, all the rating revisions that we have made with this report are summarised. In our overall coverage universe, we revisited valuations for 49 companies (Figure 46 on page 23). Consequently, 32 out of 62 companies that we cover are Buy rated, while the number of Hold and Sell rated companies are 28 and 2, respectively.

Figure 1: Rating revisions Upgrades Downgrades From Hold to Buy From Sell to Hold From Buy to Hold From Hold to Sell Aygaz Brisa Akcansa Anadolu Sigorta Celebi GH Turkish Airlines Bim Coca-Cola Icecek Cimsa Odas DO & CO Yazicilar Holding Dogan Holding Gubretas KordSA Pinar Sut Soda Sanayii Trakya Cam Source: UNLU & Co estimates

Top picks for 2017

Based on our macro outlook for 2017, we identify five main investment themes that could play out well in 2017. We discuss each of these themes in detail and determine 12 preferred names around the themes mentioned below (see page 27).

1. Banks outshine non-financials from a relative risk/reward perspective 2. Hedge currency risk through exporters and FX earners 3. High dividend payers should outperform, at least during the dividend season 4. Seeking recovery and restructuring stories 5. Generating alpha through quality small-caps

Among these 12 ideas, we name the most liquid ones as our Top Picks for 2017. Accordingly, Akbank, Garanti, Coca-Cola Icecek, Enka Insaat, Tupras, Turk Traktor and Ulker Biskuvi are the constituents of our model portfolio (see pages 35-43).

Equity Research 4

Figure 2: Top picks for 2017

Mcap Current Target Company Name Ticker Rec. Upside Comments (TLm) Price (TL) Price (TL) Akbank has superior operational metrics such as above-sector RoE, below-sector LDR, better cost Akbank AKBNK 31,360 BUY 7.84 9.40 20% control vs peers, strong capitalisation and better CoR evolution with full NPL coverage. Garanti stands out with its above-sector RoE and capital ratios. The bank also trades at very attractive Garanti Bank GARAN 32,802 BUY 7.81 9.75 25% multiples compared to its own history and global peers. CCI has been a major underperformer since the beginning of 2014 due to the headwinds in both Coca Cola Icecek CCOLA 9,554 BUY 37.56 43.90 17% Turkey and international territories. We argue that worst is over and current valuations are attractive. Potential new contract additions and recovery in outlook for the Russian economy should be Enka Insaat ENKAI 24,444 BUY 5.82 6.92 19% supportive. Also, the company's c.USD3bn net cash makes the stock defensive in times of TL volatility. Look for 46% y/y EBITDA and 54% EPS growth in 2017. Among the prime beneficiaries of weak TL and Tupras TUPRS 19,533 BUY 78.00 91.00 17% recent natural gas tariff cut. Has best in class FCF yield, ROE & ROIC profile among blue-chips. TL depreciation and import duty tax would create headwinds for importers, which is positive for Turk Turk Traktor TTRAK 4,213 BUY 78.95 92.00 17% Traktor. Looking for 16% y/y EBITDA growth in 2017. Dividend yield (8%) is attractive. The stock trades at historical low multiples. Potentially solid 4Q16 resulta should support to reinstate Ulker Biskuvi ULKER 5,681 BUY 16.61 22.30 34% investor confidence. Long term growth story with the pladis structure remain unchanged. Source: UNLU & Co analysis

Equity Research 5

Macroeconomic outlook

With US economic growth picking up from its subdued pace, we forecast three rate The secular shift in global capital hikes from the Fed in 25bps instalments, in line with the market consensus. The inflows towards developed markets consequent secular shift in global capital inflows towards developed markets requires requires solid policy-making to solid policy-making to restore economic stability in Turkey. The current downside restore economic stability in Turkey risks to the Turkish economy include a sustained rally in energy commodities markets, geopolitical risks with terrorism hitting touristic areas, the continuing purge of the failed coup-plotters, the rising likelihood of a public referendum in 1H17 and Fitch lowering Turkey’s sovereign credit rating one notch to non-investment grade.

On the backdrop of local and global risks, our underlying assumptions for 2017 consist of some stabilisation in the political scene after the expected constitutional referendum in the first half of 2017, stabilisation in TL following the gradual interest rate hikes by CBRT and geopolitical risks declining moderately. Weak growth in 1H17, followed by recovery

In 3Q16, the Turkish economy contracted by 1.8% y/y despite high government We expect a tentative recovery in the expenditures (+24% y/y in 3Q16). The main culprit was the slump in private economic activity in 4Q16 consumption and investment, which emanated from the coup attempt and the subsequent decline in household and business confidence. We expect a tentative recovery in economic activity in 4Q16 figures amid accommodative incentives and measures. The government’s push to extend credit – mostly via public banks – and a TL250bn credit guarantee fund to boost bank loans to businesses constitute such measures.

Figure 3: Real GDP growth and private consumption growth contracted in 3Q16

Real GDP growth Private consumption growth 20%

15%

10% 4.5% 4.5% 5% -1.8% 0%

-5%

-10% 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13 3Q13 4Q13 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15 1Q16 2Q16 3Q16

Source: TurkStat Given that the recent slump was driven mostly by a sharp drop in consumption by 2.6% and in exports by 6.3%, such one-off factors should have reversed in 4Q16. But the high base from the previous year (4Q15 growth was 7.4%) and an expected negative foreign demand contribution limit a strong recovery. Risks are tilted to the downside for growth prospects in 4Q16 and the first half of 2017; consumer confidence slipped in November due to the TL sell-off and increased political volatility. The probability of households buying a home reached an all-time low in December. Similarly, the real estate sector’s inclination to undertake fixed investment expenditures retreated to its softest level since the 2009 recession. The decline in manufacturing PMI in December suggests weaker IP figures in the coming months. The investment appetite is likely to remain weak unless economic reforms are pursued rigorously.

Equity Research 6

Figure 4: Recent recovery in IP is an upside for growth prospects Figure 5: Consumer confidence plunged, yet to recover in 4Q16

Industrial production index, annual change % Consumer Confidence (3-m average) 20% Private consumption growth (y/y, RHS) 85 25

15% 80 20

10% 75 15 4.6 70 10 5% 2.8 69 0.2 65 5 0% 60 0

-5% 55 -5 -4.1 -3.2 50 -10 -10% -8.4 1Q11 2Q11 3Q11 4Q11 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13 3Q13 4Q13 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15 1Q16 2Q16 3Q16 4Q16 Jun-15 Jun-16 Feb-15 Feb-16 Apr-15 Apr-16 Aug-15 Oct-15 Aug-16 Oct-16 Dec-15 Source: TurkStat Source: TurkStat, UNLU & Co estimates We expect GDP to register 2.4% growth in 2016E, which implies a deceleration compared to the previous year (6% in 2015). We also expect GDP growth to remain weak until mid-2017 with 1.5% and 2% in 1Q17 and 2Q17, respectively, but the pace of growth should pick up in the second half of 2017, with 4% in 3Q17 and 3.5% in Growth is expected to remain weak 4Q17, respectively, scoring an average growth rate of 3% in 2017. We think that the in 1H17, but the pace should pick up higher government spending and decreasing political tensions after the constitutional in the later part of the year with an referendum will improve Turkey’s growth prospects. average growth rate of 3% Figure 6: Government spending’s contribution to growth is bound to remain high 6 6% 4% 4 4% 3.5% 2.0% 2 2% 2% 1.5% 0 0%

-2 -2% 4Q16 1Q17 2Q17 3Q17 4Q17 Government Spending (LHS) Domestic Consumption (LHS) Imports (LHS) Exports (LHS) GDP Growth Rate y/y (%) (RHS)

Source: UNLU & Co forecast

Unemployment rate rises significantly

Meanwhile, slower growth in 2H16 took its toll on employment: the unemployment rate rose to 11.8% in October 2016, marking the highest such rate since March 2010, when it hit 12.8%. October unemployment data also represented the sharpest rise on an annual basis since December 2014, with a 1.3pp increase. We expect the unemployment rate to have hovered around 11.8% at 2016 year-end, stretching to around 12.5% by mid-2017 and declining back to 12% levels at 2017 year-end, following economic recovery in the second half of the year.

Equity Research 7

Figure 7: Slower growth in 2H16 took its toll on employment

15% 14% 13% 12% 11.8 11% 10% 9% 11.7 8% 7% 6% Jul-08 Jul-15 Jan-05 Jan-12 Jun-11 Sep-09 Sep-16 Feb-09 Feb-16 Apr-10 Aug-05 Oct-06 Aug-12 Oct-13 Mar-06 Mar-13 May-07 May-14 Dec-07 Dec-14 Nov-10

Unemployment rate Unemployment rate (seasonally adjusted)

Source: TurkStat

Inflation is likely to surge in 1H17

Exchange rate movements, heightened A lower growth rate is bound to create less demand-pull inflation, but the exchange global uncertainty and higher oil rate movements, heightened global uncertainty and higher oil prices pose upside risks prices pose upside risks to inflation to inflation. The FX pass-through is around 10-15% in Turkey, meaning every 10% of sustained lira weakness would translate into an extra 1-1.5% headline inflation within three-six months. Contrary to CBRT’s expectations, the weakness in demand failed to limit these effects and annual inflation reached 8.5% y/y by the end of the year, significantly higher than the CBRT forecast of 7.5% for 2016. Elevated food volatility and the possibility of frequent tax and price hikes suggest headline CPI is set to remain volatile in 2017. We expect annual CPI to hover consistently above 7% y/y in 2017, and to reach 8.5% by year-end 2017, well above the CBT’s latest forecast (6.5%). We assume gradual hikes in interest rates and stabilisation of TL at lower levels for inflation to remain below 10%. In case of a dovish CBRT response and sustained depreciation in TL, inflation could reach double-digit levels in mid-2017.

Figure 8: Both headline and core inflation increased in recent Figure 9: Inflation will surge in 1H17, mainly due to FX pass- months through

14% 12-month CPI (%) UNLU & Co CPI forecasts 14.0% 12% 12.0% 10% 8.5% 10.0% 8% 8.0% 6% 7.5% 6.0% 2017E: 8.5% 4% 4.0% 2% 2.0% Jan-07 Jan-09 Jan-11 Jan-13 Jan-15 0.0% Sep-07 Sep-09 Sep-11 Sep-13 Sep-15 May-08 May-10 May-12 May-14 May-16

CPI Inflation Core Inflation: I-Index Jun-09 Jun-11 Jun-13 Jun-15 Jun-17 Feb-08 Feb-10 Feb-12 Feb-14 Feb-16 Oct-08 Oct-10 Oct-12 Oct-14 Oct-16

Source: TurkStat Source: TurkStat, UNLU & Co estimates

Equity Research 8

CBRT prefers unorthodox measures to tighten

CBRT’s reluctance to tighten CBRT’s reluctance to tighten monetary policy created sharp volatility in TL. monetary policy created sharp December’s REER weakened to 92.16, just 3% above its historical low. After volatility in TL weakening by 9% against the currency basket and by 21% versus the dollar in 2016, TL has further weakened by 4% YTD against the basket and 7% YTD versus the dollar in 2017. Most of the depreciation in the past year was a reflection of Turkey’s high sensitivity to global financial conditions and oil prices, while the recent weakness has been driven by market concerns about CBRT’s unwillingness to raise interest rates. At December’s MPC meeting, CBRT kept the one-week repo rate unchanged at 8%, while the majority of market players expected a 25bps hike. The deceleration in economic activity in 3Q16 and the gradual fall in core inflation led CBRT to pursue a “wait-and-see” policy, which translated into a loose monetary stance with the real policy rate at around 140bps (EM peers’ average is 240bps). Most problematically, the weaker the exchange rate and the higher the overshoot in inflation, the sharper the tightening is required to contain the depreciation in TL.

Figure 10: CBRT’s recent unorthodox measures pushed the average funding rate up, out of the interest rate corridor

12 11 10 9 8 7 6 Jun-16 Jun-15 Feb-16 Feb-15 Apr-16 Apr-15 Aug-16 Oct-16 Aug-15 Oct-15 Dec-16 Dec-15 Dec-14

O/N Borrowing rate (%) O/N Lending rate (%) One-week Repo rate (%) CBRT average funding rate (%)

Source: CBRT We expect the Turkish Central Bank Before the next MPC meeting on 24th January, CBRT took some unorthodox to raise the policy rates gradually measures to control the rising trend of the euro and dollar against the lira. The bank, which funds the market with an 8% interest rate in a weekly repo auction, did not go to auction; reduced the foreign exchange reserve requirement ratios by another 50bps for all maturity brackets, reduced Banks’ borrowing limits at Interbank Money Market to TL11bn and to TL34bn at Borsa Istanbul repo markets. Instead, banks were expected to borrow at 10% from the late liquidity window, which effectively pushed the average funding cost up to around 9.1%. CBRT also decided to open the Foreign Exchange Deposits against Turkish Lira Deposits market, which aims at pushing the TL interest rates up to 10% at swap markets, and secondly at providing liquidity to banks. CBRT average funding rate to At 24th January meeting, CBRT increased the upper band of the interest rate increase with the MPC decision on corridor by 75bps to 9.25%, while it kept the lower band of the corridor unchanged January 24th at 7.25% and the weekly repo rate unchanged at 8%. On the other hand, the late liquidity window borrowing rate was increased by 100bps to 11%. Accordingly, CBRT moved the marginal funding rate upwards in line with the increasing average funding costs. We think that by CBRT would ease the use of late liquidity facility and provide the funding more heavily from the overnight window, effectively keeping the average funding cost around 9.25-9.50%. In this case, a later switch to late liquidity window can bring further tightening with the cost rising up to 10.0%. In the announcement, CBRT noted that the recovery in the economic activity is expected to continue at a moderate pace and that the aggregate demand developments support disinflation. CBRT also indicated that the Committee decided to strengthen the monetary

Equity Research 9

tightening in order to contain the deterioration in the inflation outlook, which is expected to continue in the short term due to lagged pass-through effects. Even though CBRT guided further tightening if needed, we do not think CBRT would act boldly close to the constitutional referendum (likely in early April). The number of CBRT’s Monetary Policy Meetings is to be reduced from 12 to at least eight, but the dates of meetings in 2017 are yet to be announced. However, even if an MPC meeting was to take place before the impending referendum in early spring, we expect only gradual hikes at the upper band of the corridor and the late liquidity window. We also think that the pressure on TL will diminish as political uncertainty subsides in 2H17. As a result, we expect USD/TL to stabilise in the medium term at the current levels and end the year at around 3.75.

Figure 11: December’s REER weakened to 92.16, just 3% above its Figure 12: We expect the average funding rate to further increase historical low after the interest rate hike at January’s MPC meeting

130 REER (CPI-based) 10.5 10.1 10 REER (3-years moving 120 average) 9.5

9 110 8.5 101.1 8 100 7.5 92.2 90 7 Jul-15 Jul-16 Jan-16 Jan-17 Jul-11 Sep-15 Sep-16 Jan-08 Jan-15 Jun-14 Mar-16 May-15 May-16 Nov-15 Nov-16 Sep-12 Feb-12 Apr-13 Aug-08 Oct-09 Aug-15 Oct-16 Mar-09 Mar-16 May-10 Dec-10 Nov-13

Source: CBRT Source: CBRT, UNLU & Co estimates

Current account deficit to widen in 2017

We expect the deterioration in CAD The 12-month trailing current account deficit stands at USD33.8bn in November to continue and maintain our year-end 2016, while the core deficit (overall balance less energy and gold) deteriorated forecast at USD35.4bn (4.1% of 2016E sharply and rose to USD12.4bn. Problematically, the finance account deteriorated, GDP) and at USD40.7bn for 2017 with USD1.3bn in outflows in November (USD2.7bn inflows in October), so that the (5.1% of 2017E GDP) total deficit was financed by the unidentified inflows (net errors and omissions) of USD2.7bn and a reserve drawdown of USD0.8bn. It is worrying that the current account deficit is widening despite weak economic activity and the prospect of higher energy prices, and the ongoing weakness in the tourism sector poses upside risks to the deficit. Looking forward, we think that the depreciation in TL could reduce imports, as the indicators point to an improvement in the trade deficit in December. But the upside risks lead us to expect the deterioration in CAD to continue in the upcoming months and we maintain our year-end CAD forecast at USD35.4bn (4.1% of 2016E GDP) and at USD40.7bn for 2017 (5.1% of 2017E GDP).

Equity Research 10

Figure 13: CAD is forecasted to widen and reach USD40.7bn in 2017 (5.1% of 2017E GDP)

4% 2% CAD (as % of GDP) 2% 0% 0% -2% -2% -4% -2% -4% -4% -4%-4.1% -6% -5% -5% -5.1% -6% -5% -6% -6% -8% -7%

-10% -9% 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016E 2017E

Source: TurkStat, UNLU & Co forecasts

Budget deterioration in sight, though not an immediate concern

The over-reliance on one-off non-tax The 2016 budget showed an overall deficit of TL29.3bn (1.2% of 2016E GDP) and a revenues needs to be addressed but primary surplus of TL21bn, better than the revised MTP estimates of a TL34.6bn in the near term the government has overall deficit and TL16.9bn primary surplus. However, the year-end budget indicates room to engage in growth-enhancing a significant deterioration (by 29.4% y/y) compared to a TL22.6bn deficit in 2015 fiscal policies year-end.

Figure 14: Budget deficit increases in 2017, but remains sound in historical standards

8 6 4 1.3 0.9 1.0 2 0 -2 -4 -1.0 -1.2 -1.3 -6 -8 -10 -12 -14 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017E

Primary balance as % of GDP Budget balance as a % of GDP

Source: Ministry of Finance, UNLU & Co forecasts The increases in military expenditure and government spending in the face of slower economic activity were the main drivers of the weak budget performance in 2016. The yearly budget worsened despite the one-off tax revenues stemming from the tax amnesty, the third such amnesty after 2011 and 2014. Among TL77bn of applications for the restructuring of debt in 2016, with a collection ratio of 50% (average of 2011 and 2014), the total collection is expected to be TL38bn. Given that TL12.3bn was collected in 2016, the remaining TL25.5bn should be collected in the next two years, as payments can be made in 36 instalments. These are not included in the 2017 budget, so they will support an expansionary budget in the face of sluggish growth. So even if the increasing expenditures and over-reliance on one-off non-tax revenues needs to be addressed in the medium term, in the near term the government has room to engage in growth-enhancing fiscal policies. Accordingly, we forecast the budget deficit/GDP ratio at 1.3% for 2017.

Equity Research 11

Figure 15: Budget revenues stagnate whereas expenditures increase

20% 12m-trailing, y/y change 18% 16% 15% 14% 12% 10% 8% 6% Jul-13 Jul-14 Jul-15 Jul-16 Jan-13 Jan-14 Jan-15 Jan-16 Apr-13 Apr-14 Apr-15 Apr-16 Oct-13 Oct-14 Oct-15 Oct-16

Budget expenditures Budget revenues

Source: Ministry of Finance

The highly leveraged private sector Private sector is highly leveraged suffers from a feedback loop where Turkish corporates have a net FX debt of USD212bn, equal to a sizeable 30% of currency weakness results in FX GDP. These primarily dollar liabilities are largely unhedged and a macro-prudential accumulation and further weakness of framework to manage the private sector’s foreign exchange risks, though rumoured, TL is yet to be launched. This results in a feedback loop where a currency weakness results in FX accumulation and further weakness of TL. The external debt of the Turkish banks and corporates adds up to USD300bn, constituting around 71% of all debt.

Figure 16: Private sector external debt is visibly increasing Figure 17: Private sector has a solid track record to roll over their debts

External Debt, (USD bn) 250% 138 150 140 200% 135 130 150% 130 110 125 100% 90 120 50% 115 70 110 0% 50 105 Jul-15 30 100 Jul-16 Jan-15 Jan-16 Sep-15 Sep-16 Mar-15 Mar-16 May-15 May-16 Nov-15 Nov-16

2Q11 4Q11 2Q12 4Q12 2Q13 4Q13 2Q14 4Q14 2Q15 4Q15 2Q16 Corporate Borrowing Rollover Ratio (12-m average) Banking Borrowing RR (12-m average) Banks Public Non-Financials Public Borrowing RR (12-m average) Source: CBRT Source: CBRT The ongoing depreciation pressure in TL weighs negatively on leverage metrics and also eats up the equity base of the companies via FX losses. The most heavily FX- indebted sectors are energy, transportation and construction, but the state guarantee on loans for high-cost mega-projects (USD31bn according to CBRT’s report) in these sectors provides a backup against exchange rate and demand shocks. Although we expect external financing not to be a major concern in 2017 thanks to the sizeable foreign currency borrowings and the private sector’s track record to roll over its debts, the non-financial private sector leverage constitutes a systemic risk for Turkey, with one-third of the listed companies having a leverage ratio (net debt/EBITDA) of more than 5x according to our calculations. The aggregate net debt to EBITDA in the system today is 2.3x vs 1.1x in 2010 (and 0.3x in 2003), despite weak investment appetite over the past couple of years. We continue to consider 2.5x net debt/EBITDA as a stressful level for most companies, given that the average maturity

Equity Research 12

on a corporate loan is around three years. We do not expect an immediate improvement in this picture and forecast the private sector debt to remain at 31% of 2017E GDP.

Figure 18: Corporate sector open position must be addressed in the medium term

250 35% 30% 200 25% 150 20% 100 15% 10% 50 5% 0 0% 1Q07 3Q07 1Q08 3Q08 1Q09 3Q09 1Q10 3Q10 1Q11 3Q11 1Q12 3Q12 1Q13 3Q13 1Q14 3Q14 1Q15 3Q15 1Q16 3Q16

Corporate open position (USD bn) Corporate Sector Open Position (as % of 4-quarter rolling GDP)

Source: CBRT

Political risks weigh on the economy

The utter unpredictability of politics The political scene is, in short, chaotic. The utter unpredictability of politics repels repels investors and contributes to investors and contributes to the volatility of the currency. Turkey's Parliament the volatility of the currency recently concluded the first reading and approved the draft constitutional reforms with 340 votes in favour to 135 against. The amendment package proposes a set of constitutional changes, including a switch to a presidential system of governance, grants the President sole authority to issue decrees, declare states of emergency, rule the country with resolutions during states of emergency and appoint public officials as well as half of the highest judges. It’s expected that the bill will garner enough votes in Parliament by the end of January and that the referendum will be held on 26 March at the earliest, or 2 April at the latest. Currently, the surveys seem to indicate a possible win for Erdogan; in a December survey by MetroPoll, 46.7% of participants said they supported a presidential system, while 41.0% opposed it and 12.3% were undecided. Our base scenario is a “yes” vote in the referendum, which would grant political and economic stability to Turkey at least in the short term. We are hopeful that in such a case, the long-sought after reforms could be implemented more decisively and swiftly once the economy becomes the priority. The diminishing political tensions would also benefit business and consumer confidence. On the other hand, the further centralisation of power could put further pressure on the opposition as well as the CBRT, whose independence is already thought to be compromised. However, a “no” vote would only prolong the political uncertainty and result in an early election sometime in the spring, when AKP is likely to form the new government.

Equity Research 13

Fixed income market outlook Turkey could lose its last IG rating from Fitch Ratings

On 23 September, 2016, Moody’s downgraded its sovereign credit rating on Turkey by one notch to ‘Ba1’ from ‘Baa3’, following a two-month review after the attempted coup on 15 July. After the Moody’s downgrade, Turkey has lost its investment grade status and currently Turkey has an investment grade rating only by Fitch Ratings. Moody’s downgraded its sovereign In accordance with EU regulation, both the S&P and Fitch rating agencies released credit rating on Turkey by one notch their 2017 sovereign rating release calendars for Turkey in December 2016. We note to ‘Ba1’ on 23 September, 2016 that deviations from this calendar are possible, particularly if there are ad hoc developments that materially affect Turkey’s sovereign credit quality, as we experienced after the attempted coup in July. In 2017, the first credit review will be on 27 January by Fitch, followed by S&P on 5 May. We note that Moody’s did not designate any rating review dates for Turkey, since Turkey is not covered by a lead analyst based in the EU. Fitch has already emphasised that political instability, lack of fiscal discipline, and a worsening in external imbalances could cost Turkey its last investment grade rating. The fiscal discipline and the external balances should not be a concern, at least in the short term, but we expect the political tensions to escalate further before diminishing after the constitutional referendum. If, in line with our expectation, CBRT does not respond to the market’s demand for a substantial hike and prefers a step-by-step hike, we expect Fitch to downgrade Turkey’s rating.

Figure 19: Dates of Turkey’s sovereign credit reviews

Fitch Moody's S&P A- A3 A-

BBB+ Baa1 BBB+

Grade BBB Baa2 BBB

Investment Investment BBB- Baa3 BBB-

BB+ Ba1 BB+ BB Ba2 BB

Grade BB- Ba3 BB-

Speculative B+ B1 B+ Outlook NEGATIVE STABLE STABLE 27-Jan-17 05-May-17 Review Dates 21-Jul-17 03-Nov-17

Source: Fitch Ratings, Moody’s, S&P Global Ratings In 2017, the first credit review will be If Fitch Ratings downgrades Turkey’s credit rating, there will be no index implications on 27 January by Fitch, followed by S&P for Turkey’s sovereign bonds since Turkey has already lost its IG status after the on 5 May Moody’s downgrade, and we think that the downgrade risk is largely priced into markets. However, losing the last IG rating would translate into higher risk weightings for Turkish banks’ FX exposure, adversely affecting the regulatory capital ratios of Turkish banks.

Equity Research 14

CDS increase sharply after the attempted coup

The Moody’s downgrade caused Turkey’s five-year CDS spread to widen by 15bps to 262bps, from 247bps, on the first trading day and increase by 64bps within one week. Due to continuation of geopolitical However, CDS spreads recovered within two weeks, returning to the levels seen risks and political uncertainties, we do before the downgrade. But due to increasing political uncertainties in Turkey and not expect a decrease in CDS in 2017 other global developments, such as the US presidential election outcome, Turkey’s CDS spread had widened once again, to 273bps, as of the end of December.

Figure 20: Turkey’s five-year CDS remains steady compared to the beginning of 2016

340 After the attempted 320 coup, Turkey's 5-yr CDS widened by 64bps 285 300 within one week

280

260 273 240

220

200 Jul-16 Jul-16 Jan-16 Jan-16 Jan-17 Jun-16 Sep-16 Sep-16 Feb-16 Apr-16 Aug-16 Oct-16 Mar-16 Mar-16 May-16 May-16 Dec-16 Dec-16 Nov-16 Source: Bloomberg Due to the continuation of political uncertainties and geopolitical risks in Turkey, we do not expect a decrease in Turkey’s five -year CDS in 2017. Sovereign USD bond yields increase by c.80bps in 2016

Turkey’s USD sovereign yield curve shifted upwards by c.80bps in 2016, mainly due to the increase in political uncertainties, geopolitical risks and the Fed rate hike We expect further upward shifts in the (25bps). At the beginning of 2016, the yields of Eurobonds with maturities of six-10 sovereign USD bond yields in 2017 years were at 4.5-5.0%, respectively, and currently they are at 5.3-5.8%. Due to our base-case scenario, which is the continuation of Fed rate hikes reaching the 1.25- 1.50% range at the end of 2017, we do not expect the sovereign USD bond yield curve to shift downwards in 2017, on the contrary, a further upward shift is more likely in the yield curve.

Figure 21: USD Turkey sovereign curve shifted upwards in 2016

7% 31/12/2016 6% 01/01/2016 5%

4%

3%

2%

1%

0% 1Y 5Y 10Y 15Y 20Y 25Y 30Y

Source: Bloomberg

Equity Research 15

Equity market outlook Current valuations reflect a crisis scenario

Turkish equities have been trading sideways for a prolonged period of time. It seems that there is a thin 20% trading range as the headline BIST 100 Index has oscillated between 70k and 86k 87% of the time since late 2012. We believe this performance was purely driven by the global risk-on/off mood during this time frame as there was limited support from local politics and macro. Turkish equities trade at an 8.3x P/E, Needless to say, Turkish equities have de-rated substantially over the course of the implying a 31% discount vs its peer past three years (since the taper tantrum) and the P/E discount versus its EM group counterparts climbed to historical highs. Currently, Turkish equities trade at an 8.3x P/E multiple, implying a 31% discount vs the benchmark MSCI EM Index, the largest gap measured since 2009, when the global financial crisis broke out.

Figure 22: BIST 100 Index P/E (x) multiple yearly averages Figure 23: BIST 100 vs MSCI EM P/E (x) discount over the years

10.1 9.8 9.8 -2% 9.6 -6% 9.4 9.3 -9% -9% -15% -17% 8.3 8.2 8.0 -30% -33% -33% 2009 2010 2011 2012 2013 2014 2015 2016 2017 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates The depressed level of multiples shows that investor confidence towards Turkish Turkish Equities offer a solid assets has crumbled in the past few years. In fact, they treat Turkey as a crisis risk/reward profile candidate, which we think is certainly unwarranted despite all the negative political/economic developments. We believe Turkey offers a solid risk/reward profile at the current depressed levels.

Figure 24: BIST 100 Index vs MSCI EM P/E (x) multiple Figure 25: BIST 100 vs MSCI EM P/E (x) discount over the years

13.0 20%

12.0 10%

11.0 0%

10.0 -10%

9.0 -20%

8.0 -30% Prem / disc of BIST-100 vs. MSCI EM BIST-100 Index MSCI EM Index historical average 7.0 -40% Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

Equity Research 16

Currently, Turkey trades at a deep discount vs major EM markets other than Russia (though the premium is also narrowing). As illustrated below, Turkey tends to trade in line with Brazil, South Africa and Poland as far as P/E multiples are concerned. Yet, this correlation was broken in mid-2013 with rising political concerns.

Figure 26: Turkey vs Brazil P/E (x) multiple Figure 27: Turkey vs Mexico P/E (x) multiple

Turkey Brazil Turkey Mexico 15.0 20.0 14.0 18.0 13.0 16.0 12.0 14.0 11.0 12.0 10.0 9.0 10.0 8.0 8.0 7.0 6.0 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

Figure 28: Turkey vs Russia P/E (x) multiple Figure 29: Turkey vs Poland P/E (x) multiple

Turkey Russia Turkey Poland 12.0 15.0 11.0 14.0 10.0 13.0 9.0 12.0 8.0 11.0 7.0 10.0 6.0 9.0 5.0 8.0 4.0 7.0 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

Figure 30: Turkey vs S. Africa P/E (x) multiple Figure 31: Turkey vs Asia P/E (x) multiple

Turkey S. Africa Turkey Asia Median 18.0 16.0

16.0 14.5

14.0 13.0

12.0 11.5

10.0 10.0

8.0 8.5

6.0 7.0 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

Equity Research 17

A decline in inflation expectations Looking at the P/B-ROE matrix of the EM universe, which we believe is the best way should lead to a re-rating to identify valuation and profitability variation among countries, Turkey stands out as one of the most attractive markets. However, we think this does not fully capture the performance drivers of the market. In Figure 33 below, we show the same matrix with adjusted ROE figures for inflation expectations. Turkish equity valuations still look attractive in this exercise, considering the expected normalisation in inflation and higher ROE assumptions for the banks. In fact we see scope for a re-rating of 20% — that is almost the same upside that we arrive at with our bottom-up index target of 96k.

Figure 32: P/B vs ROE matrix in EM Figure 33: P/B vs ROE (inflation adjusted) matrix in EM 3.0 3.0

India 2.5 India 2.5 Mexico Mexico Philippines Indonesia 2.0 2.0 Philippines Indonesia Thailand Malaysia Thailand Brazil Turkey China 1.5 Chile Malaysia China Taiwan 1.5 Target Taiwan Brazil Czech R.

P/B (2017E) Hungary Poland Hungary P/B (2017E) Chile Czech R. 1.0 Turkey 1.0 Poland Turkey S. Korea Russia S. Korea Current Russia 0.5 0.5 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 Stated ROE (2017E) Inflation adjusted ROE (2017E)

Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

De-rating has primarily been driven by banks

Turkish banks have been trading at a deep discount versus their non-financial peers for a long time. We believe the discount is structural to a great extent, as Turkish banks are not able to generate economic profit, which we measure as the difference between ROE and COE. However, the level of the discount increased substantially in the past two years, which is very hard to justify for us.

Figure 34: Banks vs non-banks P/E (x) multiple Figure 35: Banks vs non-banks P/E (x) discount over the years 0% Banks Non-banks 13.0 -5% 12.0 -10% 11.0 -15% 10.0 9.0 -20% 8.0 -25% 7.0 -30% 6.0 -35% 5.0 -40% Jul-12 Jan-15 Jan-10 Jun-15 Jun-10 Sep-16 Sep-11 Feb-12 Apr-16 Apr-11 Oct-13 Aug-14 Mar-14 May-13 Dec-12 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Nov-15 Nov-10 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

Equity Research 18

Turkish banks’ index weight declined The annual development of multiples shows that there is a clear composition change to 30% from 50% five years ago within the equity space in favour of non-banks. Regardless of the fundamental reasons (such as earnings performance), the banks’ multiples are on a declining trend. Non- banks’ on the other hand trade in a very narrow P/E band of 9-11x. We note that Turkish’ banks weighting in the benchmark BIST 100 Index also decreased to as low as 30%, from close to 50% at the end of 2009, following this dislocation.

Figure 36: Banks and non-banks P/E (x) multiple yearly averages Figure 37: Banks vs non-banks P/E (x) discount over the years

Banks Non-banks 11.1 10.8 10.7 10.4

10.1 -9% 9.9 -12% 9.2 9.2 9.0 -15% 8.7 8.7 -16% 8.4 8.4 8.0 7.9 -22% 7.2 -26%

6.0 -32%

5.4 -35% -40% 2009 2010 2011 2012 2013 2014 2015 2016 2017 2009 2010 2011 2012 2013 2014 2015 2016 2017

Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

The value discount of non-banks is also on the rise

Having stressed that Turkish equities’ underperformance is primarily driven by the banks, currently non-banks also trade at a deep PE discount vs their respective peer groups (Figure 38). Although the discount on EV/EBITDA multiple has not changed over the past five years, the P/E discount is an increasing trend. We think the main reason behind this is diminishing returns on new investments. Increasing depreciation charges, coupled with financial expenses, have offset the EBITDA improvement in non-banks over the past years and led to weaker structural EPS prospects.

Despite this, we believe there are still number of attractive themes within the non- financial space as highlighted in detail in our Top Picks section.

Figure 38: Turkish non-banks P/E (x) vs peer group Figure 39: Turkish non-banks EV/EBITDA (x) vs peer group

Turkey Non-Financials Peer Group Median Turkey Peer Group Median 16.0 10.0

15.0 9.0 14.0 8.0 13.0 12.0 7.0 11.0 6.0 10.0 5.0 9.0 8.0 4.0 7.0 3.0 Jul-10 Jul-10 Jul-11 Jul-11 Jul-12 Jul-12 Jul-13 Jul-13 Jul-14 Jul-14 Jul-15 Jul-15 Jul-16 Jul-16 Jan-10 Jan-10 Jan-11 Jan-11 Jan-12 Jan-12 Jan-13 Jan-13 Jan-14 Jan-14 Jan-15 Jan-15 Jan-16 Jan-16 Apr-10 Apr-10 Apr-11 Apr-11 Apr-12 Apr-12 Apr-13 Apr-13 Apr-14 Apr-14 Apr-15 Apr-15 Apr-16 Apr-16 Oct-10 Oct-11 Oct-12 Oct-13 Oct-14 Oct-15 Oct-16 Oct-10 Oct-11 Oct-12 Oct-13 Oct-14 Oct-15 Oct-16

Source: Bloomberg, UNLU & Co estimates Source: Bloomberg, UNLU & Co estimates

Equity Research 19

Key revisions and rating changes We revisit our entire coverage universe’s valuation

Rates likely to remain higher in the coming cycle

Based on Bloomberg consensus, the US 2016 was a year of surprises for the world, especially the outcome of the Brexit vote 10-year yield could remain on an and the US elections. The latter, in particular, could have a prolonged impact on the increasing path at least for six quarters global economy and interest rates, in our view. A Republican-controlled Congress and White House are likely to push for fiscal stimulus to boost the US economy. On the other hand, the Fed signalled a tighter monetary policy in 2017 with three potential rate hikes, instead of two. If fiscal spending in the US tightens the labour market and leads to wage growth-driven inflationary pressures, Fed chair Janet Yellen may be forced to act even faster. While calling US rates is beyond the scope of this report, we can easily suggest that near-term trends suggest a higher interest rate environment globally. The US 10-year bond yield, which was in a downtrend until 3Q16, jumped almost 100bps following the US election and the Fed’s recent statement. Based on Bloomberg consensus, the US 10-year yield could remain on an increasing path at least for six quarters.

Figure 40: US 10-year yield is expected to continue to rise

3.5%

3.0%

2.5%

2.0%

1.5%

1.0% Jul-16 Jan-14 Jan-17 Jun-18 Jun-14 Jun-17 Sep-14 Sep-17 Feb-16 Apr-17 Aug-15 Oct-16 Mar-15 Mar-18 Mar-14 May-15 May-16 Dec-14 Dec-17 Nov-15 Source: Bloomberg consensus

Turkey is not likely to decouple

Interest rates in Turkey remained relatively stickier throughout 2016 following a c.300bps increase in 2015. Macro and political shocks and weaker TL have weighed negatively on TL interest rates in the past two years.

Figure 41: Turkey 10-year yield remained sticky in 2016 following a jump in 2015 12.5%

11.5%

10.5%

9.5%

8.5%

7.5%

6.5% Jan-14 Jan-15 Jan-16 Jan-17 TL 10-yr government bond yield 200 days MA

Source: Bloomberg

Equity Research 20

We revise up our RFR assumptions to Considering global trends and the macro outlook in Turkey, we do not expect TL 10.5% assuming that rates will remain interest rates to decline in the near term. So far, we have used 9.5% as our base-case elevated RFR assumption in our valuation, which is mainly based on the mid-cycle average of 10-year generic TL rates. Although the 200 days moving average interest rate is 10%, we revise up our RFR assumptions to 10.5% assuming that rates will remain elevated. We do not expect 10-year TL yields to remain around the 11.0-11.5% level on a sustainable basis. We expect CBT to continue to take the necessary steps to stabilise currency moves and maintain its hawkish stance. Therefore, the long end of the yield curve is likely to come down to below 11% in the coming months. Based on our updated macro forecast, the next three years’ average 10-year yield is likely to be 10.5%, which is why we set our RFR assumption at 10.5%.

Figure 42: We expect the next three-year average interest rate to be 10.5% 12.0% 11.5% 11.0% 10.5% 10.0% 9.5% 9.0% 8.5% Jul-16 Dec-16 May-17 Oct-17 Mar-18 Aug-18 Jan-19 Jun-19 Nov-19 TL 10-yr government bond yield 200 days MA Past 3-yr average Next 3-yr average Source: UNLU & Co

Revised COE leads to an 11% average cut in our TPs

The weighted average impact of the With this report and recently published banking and telecom sector reports, we have 100bps COE increase is -15% for our revisited the valuation models for our entire coverage universe and incorporated our one-year forward looking index target new macro and COE assumptions. A 100bps increase in our COE assumption has led to an 11% decline in our TPs, on average. For heavy weight banks, the impact of COE revision is as high as 19%. Therefore, the weighted average impact of the 100bps COE increase is -15% for our one-year forward looking index target.

Figure 43: Impact of a higher COE on our TPs differs from sector to sector

40% Free float MCAP weight Average impact of 100bps higher COE on TPs 30% 20% 10% 0% -19% -10% -6% -13% -15% -11% -9% -11% -4% -11% -9% -13% -10% -16% -10% -20% -30% TMT Banks Utilities Aviation Defence Retailers Cements Insurance Real Estate Commodities Conglomerates White Goods Autos & Partsand Autos & Glass materials and Food and Beverages Source: UNLU & Co estimates

Equity Research 21

Our bottom-up BIST 100 index target is 96,350

We view relatively more attractive Despite the negative impact of a higher COE assumption, our bottom-up index target upside potentials in the banks, real for the next 12 months is 96,350, which offers 16% upside potential. While this may estate and conglomerates sectors not look attractive, we view relatively more attractive upside potentials in the banks, real estate and conglomerates sectors. On the other hand, the retail, TMT, automotive, white goods, cement, aviation, glass and defence sectors offer relatively less attractive upsides.

Figure 44: Weighted average sectoral upsides

25% Bottom-up index upside: 20% 16% 15%

10%

5%

0% TMT Banks Utilities Aviation Defence Retailers Cements Insurance Real Estate Commodities Conglomerates White Goods White Autos & PartsAutos & and Glass and materials Food and Beverages Food and Source: UNLU & Co estimates As previously mentioned, the 100bps higher COE resulted in a 15% weighted average decline in our TPs and with a 9.5% RFR instead of 10.5%, our bottom-up index target would increase to 111,000, which offers 34% upside from the current BIST 100 level. We change the ratings of 18 companies

32 out of 62 companies that we cover We upgrade five companies from Hold to Buy and two companies from Sell to Hold, are Buy rated, while the number of on the other hand, we downgrade 10 companies from Buy to Hold and one company Hold and Sell rated companies are 28 from Hold to Sell due to relatively limited upside potentials. In Figure 45, all the rating and 2, respectively revisions that we have made with this report are summarised. In our overall coverage universe, we revisited valuations for 49 companies (Figure 46). Consequently, 32 out of 62 companies that we cover are Buy rated, while the number of Hold and Sell rated companies are 28 and 2, respectively.

Figure 45: Rating revisions Upgrades Downgrades From Hold to Buy From Sell to Hold From Buy to Hold From Hold to Sell Aygaz Brisa Akcansa Anadolu Sigorta Celebi GH Turkish Airlines Bim Coca-Cola Icecek Cimsa Odas DO & CO Yazicilar Holding Dogan Holding Gubretas KordSA Pinar Sut Soda Sanayii Trakya Cam Source: UNLU & Co estimates

Equity Research 22

Figure 46: Key revisions for UNLU & Co Turkey coverage universe Price (TL) TP (TL) Rating Company Name Ticker as of 23 Jan Upside Old New Change Old New Change Banks Akbank AKBNK.IS 7.84 20% 9.40 9.40 unch. Buy Buy Albaraka ALBRK.IS 1.24 13% 1.40 1.40 unch. Hold Hold Garanti Bank GARAN.IS 7.81 25% 9.75 9.75 unch. Buy Buy Halkbank HALKB.IS 10.58 26% 13.30 13.30 unch. Buy Buy Isbank ISCTR.IS 5.57 25% 6.95 6.95 unch. Buy Buy TSKB TSKB.IS 1.42 34% 1.90 1.90 unch. Buy Buy Vakifbank VAKBN.IS 4.66 27% 5.90 5.90 unch. Buy Buy Yapi Kredi YKBNK.IS 3.49 15% 4.00 4.00 unch. Hold Hold Insurance Aksigorta AKGRT.IS 2.69 12% 1.86 3.00 +61% Hold Hold Anadolu Hayat ANHYT.IS 5.10 37% 6.44 7.00 +9% Buy Buy Anadolu Sigorta ANSGR.IS 2.12 -1% 1.62 2.10 +30% Hold Sell AvivaSA AVISA.IS 20.86 15% 24.00 24.00 unch. Buy Buy TMT Logo LOGO.IS 49.60 43% 68.00 71.00 +4% Buy Buy Turk Telekom TTKOM.IS 5.62 10% 6.20 6.20 unch. Hold Hold Turkcell TCELL.IS 10.87 0% 10.90 10.90 unch. Hold Hold Conglomerates Dogan Holding DOHOL.IS 0.83 4% 0.76 0.86 +13% Buy Hold Enka Insaat ENKAI.IS 5.82 19% 5.70 6.92 +21% Buy Buy Gozde Girisim GOZDE.IS 1.90 25% 3.18 2.37 -25% Buy Buy Koc Holding KCHOL.IS 14.62 18% 16.80 17.20 +2% Buy Buy Sabanci Holding SAHOL.IS 9.51 20% 11.20 11.40 +2% Buy Buy Tekfen Holding TKFEN.IS 7.12 38% 9.10 9.80 +8% Buy Buy Yazicilar Holding YAZIC.IS 14.32 28% 20.00 18.30 -9% Hold Buy Commodities Aygaz AYGAZ.IS 12.30 18% 13.80 14.50 +5% Hold Buy Gubre Fabrikalari GUBRF.IS 4.88 9% 7.20 5.30 -26% Buy Hold Erdemir EREGL.IS 5.61 16% 5.10 6.50 +27% Hold Hold Kardemir KRDMD.IS 1.21 20% 1.70 1.45 -15% Buy Buy Petkim PETKM.IS 4.08 10% 5.10 4.50 -12% Hold Hold Tupras TUPRS.IS 78.00 17% 89.00 91.00 +2% Buy Buy Autos & Parts and White Goods Arcelik ARCLK.IS 22.48 -1% 20.90 22.20 +6% Hold Hold Brisa BRISA.IS 6.66 1% 7.10 6.70 -6% Sell Hold Dogus Otomotiv DOAS.IS 9.04 8% 9.50 9.80 +3% Hold Hold Ford Otosan FROTO.IS 32.92 12% 36.30 36.80 +1% Buy Buy KordSA KORDS.IS 6.96 7% 5.70 7.45 +31% Buy Hold Tofas TOASO.IS 25.38 12% 27.80 28.50 +3% Buy Buy Turk Traktor TTRAK.IS 78.95 17% 94.00 92.00 -2% Buy Buy Defence Aselsan ASELS.IS 13.70 10% 9.00 15.10 +68% Hold Hold Real Estate Emlak GYO EKGYO.IS 3.04 18% 3.80 3.60 -5% Buy Buy IS GYO ISGYO.IS 1.53 31% 2.20 2.00 -9% Buy Buy Torunlar GYO TRGYO.IS 4.21 16% 5.70 4.90 -14% Hold Hold Retailers Bim BIMAS.IS 53.70 -1% 62.50 52.90 -15% Buy Hold Bizim Toptan BIZIM.IS 11.84 31% 21.50 15.50 -28% Buy Buy Migros MGROS.IS 18.23 19% 23.20 21.70 -6% Buy Buy Teknosa TKNSA.IS 3.75 -1% 6.90 3.70 -46% Hold Hold Cements Akcansa AKCNS.IS 14.30 7% 16.00 15.30 -4% Buy Hold Cimsa CIMSA.IS 17.36 9% 18.00 19.00 +6% Buy Hold Glass and materials Anadolu Cam ANACM.IS 3.07 12% 2.45 3.45 +41% Hold Hold Sise Cam SISE.IS 3.97 7% 3.65 4.25 +16% Hold Hold Soda Sanayii SODA.IS 5.77 13% 5.46 6.50 +19% Buy Hold Trakya Cam TRKCM.IS 2.93 13% 3.10 3.30 +6% Buy Hold Aviation Celebi GH CLEBI.IS 25.18 19% 38.00 30.00 -21% Hold Buy DO & CO DOCO.IS 236.90 10% 400.00 261.00 -35% Buy Hold Pegasus Airlines PGSUS.IS 16.00 10% 18.90 17.60 -7% Hold Hold TAV Airports TAVHL.IS 15.35 19% 23.00 18.20 -21% Buy Buy Turkish Airlines THYAO.IS 5.20 2% 7.00 5.30 -24% Sell Hold Food and Beverages Anadolu Efes AEFES.IS 21.54 6% 21.30 22.80 +7% Hold Hold Coca Cola Icecek CCOLA.IS 37.56 17% 40.00 43.90 +10% Hold Buy Pinar Sut PNSUT.IS 16.65 15% 23.40 19.10 -18% Buy Hold Tat Gida TATGD.IS 6.21 19% 7.60 7.40 -3% Buy Buy Ulker Biskuvi ULKER.IS 16.61 34% 22.30 22.30 unch. Buy Buy Utilities Ak Enerji AKENR.IS 0.85 -18% 1.07 0.70 -35% Sell Sell Aksa Enerji AKSEN.IS 3.05 25% 3.80 3.80 unch. Buy Buy Odas Elektrik ODAS.IS 8.90 29% 8.40 11.50 +37% Hold Buy Source: UNLU & Co estimates

Equity Research 23

Consensus revision trends Earnings revisions favour the banks

Since September, the cumulative EPS revisions for Turkish equities were broadly stable (Figure 47). However, there was a clear divergence in earnings revisions for banks and non-banks. Figure 48 shows that in the past two quarters, banks have been subject to notable upward earnings revisions while non-banks were downgraded massively on the back of increasing FX translation losses and EBITDA downgrades.

Figure 47: Monthly EPS revisions and 3M moving average trend Figure 48: Quarterly EPS revisions for banks and non-banks line for Turkish equities

3.0% Banks Non-banks 2.0% 20% 16% 14% 1.0% 15% 10% 0.0% 5% 4% 4%

5% 3% 2% 2% 2%

-1.0% 0% 0% -2.0% -5% 0% -1% -2% -2% -2% -3% -3% -3%

-3.0% -4% -4%

-10% -5% -6% -4.0% -15% -10% -5.0% -20%

-6.0% -25% -18% Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15 1Q16 2Q16 3Q16 4Q16 Apr-12 Apr-13 Apr-14 Apr-15 Apr-16 Oct-12 Oct-13 Oct-14 Oct-15 Oct-16 Source: Company data, UNLU & Co estimates Source: Company data, UNLU & Co estimates We think this trend will continue throughout 2017. Our updated 2017 EPS estimates for Turkish equities are only 1% above the current Bloomberg consensus expectations. However, once again, we expect the EPS revisions to diverge notably for banks and non-banks. For 2017, our earnings expectations for banks is 8% higher than Bloomberg consensus numbers. The optimistic guidance by the banks and potentially stronger-than-anticipated 4Q16 earnings are likely to be the catalyst for further upgrades. On the flipside, our earnings forecasts for non-banks are 5% lower than Bloomberg We expect consensus to raise their consensus forecasts. The TL has depreciated notably since the beginning of the year. 2017 EPS forecasts for banks by 8% We think analysts will gradually reflect this in their models for non-banks, which would push them to lower their net income forecasts throughout the year. This divergence in EPS revisions should ultimately be reflected in share price performance, which supports our bullish stance for Turkish banks.

Figure 49: Earnings deviation between UNLU & Co and Bloomberg consensus forecasts for banks and non-banks in 2016 / 2017

2016E 2017E 10% 8% 5% 5% 1% 0%

-5% -5% -6% -10%

-15%

-20% -18% Banks Non-banks Total

Source: UNLU & Co estimates, Bloomberg consensus

Equity Research 24

Looking for 25% EPS growth for TR equities

For 2017, our earnings expectation for Turkish equities implies 25% y/y growth EPS growth for Turkish equities is driven by non-banks (+47%). However, one should note that the strong EPS growth expected to accelerate in 2017, mainly for non-banks is mostly due to the favourable base effect of 2016, when the bottom- due to a reversal of FX losses line is expected to decline notably on FX translation losses. Looking at the cumulative EPS growth for the 2016/2017 period, banks clearly stand out, with continued EPS growth.

Figure 50: EPS growth for banks and non-banks for 2016/2017

2016E 2017E 60% 47% 42% 40% 25% 20% 10% 3% 0%

-20%

-26% -40% Banks Non-banks Total

Source: UNLU & Co estimates EBITDA growth accelerates in 2017 with normalisation in some sectors

Our implied EBITDA growth for Between 2010 and 2015, the industrials in our coverage delivered 14% annual Turkish non-financials is 23% in 2017, EBITDA growth. For 2017, we expect the EBITDA growth to accelerate to 23% y/y, with normalisation in the aviation which is one of the highest levels since 2010. Once again, we should note that the sector and commodities strong growth is mainly driven by the normalisation of operational profitability of heavy weight sectors such as aviation and commodities. These two sectors were clear underperformers in 2016, with 25% EBITDA contraction on average. In 2017, we expect strong growth for both sectors, incorporating a normalisation in operational performance (Figure 52). Nevertheless, excluding both sectors, our implied 2017 EBITDA growth expectation would be 16% y/y, which is still above the historical average. Overall, we think operational performance will improve on a y/y basis in 2017 for non-banks. However, the earnings outlook is still better for banks compared to non- banks.

Figure 51: EBITDA growth evolution Figure 52: 2017 EBITDA growth on a sectoral basis

30% 90% 77% 80% 71% 24% 25% 23% 70% 60% 20% 17% 50% 15% 15% 40% 29% 15% 25% 12% 30% 22% 20% 19% 17% 20% 16% 10% 11% 11% 10% 10% 3% 5% 0% Glass Retail 0% REITS Utilities Cement Aviation Defence Telecom -1% -1% White goods Auto & Parts & Auto -5% Commodities Conglomerate 2010 2011 2012 2013 2014 2015 2016E 2017E Food and Beverage Food Source: UNLU & Co estimates, Bloomberg consensus Source: Company data, UNLU & Co estimates

Equity Research 25

Earnings revisions: How does Turkey stand in the EM context?

As mentioned in the previous section, the revisions to consensus’ earnings forecasts for Turkish equities have been flat in the past months. Figure 53 shows that the lack of any material index performance is also a factor for flattish forward EPS expectations.

Figure 53: One-year fwd EPS evolution of Turkish equities vs Figure 54: Relative one-year fwd EPS evolution of XU100 Index vs index performance (RHS) MSCI EM Index in USD terms vs relative index performance

XU100 performance XU100 - one year fwd looking EPS Relative index performance (XU100 vs. MSCI EM) (USD terms) 250 240 Relative EPS evoluiton (XU100 vs. MSCI EM Index) (RHS) 140 130 200 200 130 125 120 120 150 115 160 110 110 100 105 100 120 90 100 95 50 80 80 90 70 85 0 40 60 80 Feb-06 Feb-07 Feb-08 Feb-09 Feb-10 Feb-11 Feb-12 Feb-13 Feb-14 Feb-15 Feb-16 Aug-06 Aug-07 Aug-08 Aug-09 Aug-10 Aug-11 Aug-12 Aug-13 Aug-14 Aug-15 Aug-16 Apr-16 Apr-15 Apr-14 Apr-13 Apr-12 Apr-11 Aug-16 Aug-15 Aug-14 Aug-13 Aug-12 Aug-11 Dec-16 Dec-15 Dec-14 Dec-13 Dec-12 Dec-11 Dec-10 Source: Bloomberg Finance LP, UNLU & Co estimates Source: Bloomberg Finance LP, UNLU & Co estimates That said, in Figure 54, one can easily notice that Turkish equities underperformed the MSCI EM Index substantially in USD terms. This was driven by two factors: i) the weak earnings outlook in USD terms in relative context; and ii) multiple de-ratings of Turkish equities. Figure 55 puts the first reason into perspective by showing that in relative terms the EPS evolution of Turkish equities in USD terms was notably worse than that of the MSCI EM Index, especially since 2H16. Currency is also an important factor in this comparison that is worth considering. The Turkish Lira has depreciated notably against other EM countries since May 2016 due to the deterioration in Turkey’s macro outlook. However, a potential rate hike by CBT can partially offset the recent TL devaluation. This, coupled with our expectation of marginal EPS upgrades for Turkish equities, will improve the valuation of Turkish equities notably compared to the MSCI EM Index.

Figure 55: One-year forward EPS of Turkish equities and MSCI Figure 56: TL performance against EM currency basket (rebased EM Index in USD terms (indexed at Jan 2011) at Jan 2015)

XU100 Index MSCI EM Index 110 120 100 110 90 100 80 90 70 80 60 70 50 60 Jul-15 Jul-16 Jan-15 Jan-16 Jan-17 Sep-15 Sep-16 Mar-15 Mar-16 May-15 May-16 Nov-15 Nov-16 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Apr-11 Apr-12 Apr-13 Apr-14 Apr-15 Apr-16 Oct-11 Oct-12 Oct-13 Oct-14 Oct-15 Oct-16

Source: Bloomberg Finance LP, UNLU & Co estimates Source: Bloomberg Finance LP, UNLU & Co estimates

Equity Research 26

Key themes for 2017

We see five main investment themes that are set to play out well in 2017. These are:

1. Banks outshine non-financials from a relative risk/reward perspective.

2. Hedge currency risk through exporters and FX earners.

3. High dividend payers should outperform, at least during the dividend season.

4. Seeking recovery and restructuring stories.

5. Generating alpha through quality small-caps.

While we discuss each of these themes over the next pages of this report, the below table summarises the names that we expect to play out well in 2017. While we have a total of 12 stock ideas that we have built conviction for in 2017 around these investment themes, we trim them down to seven as our Top Picks.

Figure 57: Our stock ideas for 2017 and their related investment theme(s)

Company Name Theme 1 Theme 2 Theme 3 Theme 4 Theme 5 Top Picks Akbank X Garanti X Coca Cola Icecek X Enka Insaat X Tupras X X Turk Traktor X Ulker Biskuvi X Other ideas to play related themes TSKB X X Logo Yazilim XX Migros X Odas Elektrik XX Yazicilar Holding X

Source: UNLU & Co analysis

Equity Research 27

Theme 1: Banks outshine non-financials from a relative risk/reward perspective

We expect Turkish banks under our We remain constructive on the banking sector: Turkish banks are about to coverage to deliver 42% EPS growth demonstrate that they ended a phenomenal year in terms of earnings growth and in 2016E, followed by 10% growth in ROE expansion following the announcement of the 4Q16 financials. In fact, we expect 2017E Turkish banks under our coverage to deliver 42% EPS growth in 2016E, followed by 10% growth in 2017E. As Turkish banks failed to perform both on an absolute basis and on relative terms, banking sector valuation multiples have de-rated significantly on the back of strong earnings growth. The implied 2017 multiples look even cheaper, not only on historical terms but also relative to non-financials. As a result, we continue to prefer banks over non-financials given that banks’ multiples are far more attractive compared to those of non-banks. If CBRT hikes the rates in line with our expectations, the multiple gap may narrow in a short period of time.

Figure 58: One-year-forward P/E comparison – Xbank vs MSCI Figure 59: One-year-forward P/B comparison – Xbank vs MSCI EM banks EM banks

10.0 1.3

1.2 9.0 1.1 8.0 1.0

7.0 -33% 0.9 -30% 0.8 6.0 0.7 5.0 0.6

4.0 0.5

Xbank MSCI banks Xbank MSCI banks

Source: Bloomberg Source: Bloomberg

Turkish banks are optimistic, so are we: Turkish banks started sharing their 2017 budget guidance recently. Overall, we are of the view that Turkish banks are optimistic for 2017. Although banks expect mid-teen (13-14%) loan growth and flattish margins, improving jaws (fees-costs) and limited asset quality deterioration expectations imply double-digit (10-15%) EPS growth for 2017. We also agree that Turkish banks are capable of delivering their targets, and double-digit EPS growth is achievable for 2017. Following the changes in our latest banking report, Yapi Kredi and Albaraka Turk are rated Hold, while the rest of our coverage (Ak, Garanti, Is, Halk, Vakif, TSKB) are all rated Buy.

Figure 60: UNLU & Co banking sector assumptions

2017E Loan growth Deposit growth NII growth Fee growth Opex growth Provision growth NI growth Specific CoR AKBNK 13% 12% 15% 10% 6% 37% 1% 1.40% GARAN 12% 12% 13% 10% 6% 20% 11% 1.25% HALKB 14% 13% 16% 15% 10% 18% 28% 0.70% ISCTR 13% 13% 14% 15% 10% 39% 11% 1.20% VAKBN 13% 13% 15% 12% 10% 36% 2% 1.35% YKBNK 12% 12% 13% 10% 8% 26% 10% 1.45% ALBRK 10% 13% 10% 14% 14% 39% -2% 1.65% TSKB 16% n.a. 13% 30% 13% n.m. 10% 0.20%

Source: UNLU & Co estimates

Equity Research 28

It seems like there is a further 5% Multiples are cheaper than they seem: As mentioned above, current multiples upside to YE16 estimates, implying of Turkish banks imply that they are trading at attractive levels compared to their some 20-25% potential beat in 4Q16 peers and their own history. However, Turkish banks are even cheaper than they seem, in our view. When we compare our revised numbers with Bloomberg consensus numbers, it seems like there is further 5% upside to YE16 estimates, implying some 20-25% potential beat in 4Q16.

Figure 61: 4Q16 estimates (TL m); UNLU & Co vs Bloomberg consensus

1,400

1,200 +27% 1,000 +57% +23%

800 -22% 600 +46% +27% 400

+10% 200 -54%

0 AKBNK GARAN HALKB ISCTR VAKBN YKBNK ALBRK TSKB

Bberg implied UNLU

Source: Bloomberg, UNLU & Co estimates

Also, our cumulative 2017 numbers are 8% above that of consensus. We should also note that Turkish banks’ guidance implies higher EPS growth than our forecasts in 2017. As a result, if we assume that banks deliver their guidance, Bloomberg consensus forecasts for 2017 would need to be revised by c.15%.

Figure 62: UNLU & Co vs consensus, 2016E NI (TL m) Figure 63: UNLU & Co vs consensus, 2017E NI (TL m)

5,000 +5% 6,000 +7% +10% +13% 5,000 4,000 +1% +13%

4,000 -2% +6% +12% 3,000 +3% +11% 3,000 +9% 2,000 2,000

1,000 +3% 1,000 +2% -19% -27% 0 0 AKBNK GARAN HALKB ISCTR VAKBN YKBNK ALBRK TSKB AKBNK GARAN HALKB ISCTR VAKBN YKBNK ALBRK TSKB

Consensus UNLU Consensus UNLU

Source: UNLU & Co estimates, Bloomberg Source: UNLU & Co estimates, Bloomberg

Equity Research 29

We think asset quality deterioration Asset quality and capitalisation are two key areas of concern: The biggest will remain limited in 2017 concern for the banking sector is whether NPL ratios could once again reach 5-6%, as was the case in 2009. Despite the recent slowdown in economic activity, we think the asset quality deterioration will remain limited barring a recession in the economy. We incorporate higher COR assumptions (compared to banks’ guidance) in our models, but we believe that Turkish banks have enough of a buffer to manage rising NPL inflows in 2017.

Figure 64: NPL and specific COR evolution for the banks in our coverage (%)

6% 3.0% 5.3% 5% 2.5% 2.5% 4.1% 4% 3.8% 2.0% 3.4% 3% 1.5% 1.1% 1.2% 1.2% 2% 1.0%

1% 0.5%

0% 0.0% 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016E 2017E 2018E

NPL ratio Specific CoR

Source: UNLU & Co analysis

Every 10% TL depreciation would Another key downside risk for the sector is a sharp decline in capitalisation ratios have a negative impact of 50-60bps on due to currency devaluation and a potential downgrade by Fitch. Our findings show Tier-I banks’ CAR that every 10% TL depreciation would have a negative impact of 50-60bps on Tier-I banks’ CAR.

Figure 65: Potential impact of 10% depreciation on capital ratios

120bps 114bps 108bps 100bps

80bps

59bps 60bps 54bps 55bps 54bps 50bps 48bps 49bps 46bps 50bps 44bps 40bps 40bps 36bps 30bps 22bps 20bps

0bps AKBNK GARAN ISCTR YKBNK HALKB VAKBN TSKB ALBRK

Negative impact on Tier-I Negative impact on CAR

Source: Company financials, UNLU & Co analysis and estimates

On the other hand, the negative impact of a rating downgrade for our coverage would be c. 120bps on CAR and 105bps on Tier-I ratios. We should also note that BRSA recently announced that Turkish banks may use the Islamic International Rating Agency (which currently rates Turkey as investment grade) for calculating their risk weightings. As a result, some banks may use an alternative rating agency instead of Fitch so that there would not be any negative impact on Turkish banks’ capital ratios even if Fitch does downgrade Turkey.

Equity Research 30

Figure 66: Total impact of a potential downgrade (TL bn, unless stated otherwise)

3Q16 AKBNK GARAN ISCTR YKBNK HALKB VAKBN TSKB ALBRK Additional RW 28,185 19,727 17,385 14,792 13,655 11,567 1,022 1,904 Adjusted RWA 242,209 242,844 267,989 223,929 182,169 168,475 20,912 24,736

Adjusted CAR 13.3% 15.1% 14.8% 14.0% 12.5% 13.4% 14.5% 11.8% Adjusted Tier-I 12.3% 14.1% 12.9% 10.5% 11.6% 11.0% 13.7% 8.7%

Negative impact on CAR 175bps 133bps 103bps 99bps 101bps 99bps 74bps 98bps Negative impact on Tier-I 162bps 125bps 90bps 74bps 94bps 81bps 70bps 73bps

Source: Company financials, UNLU & Co analysis and estimates

Theme 2: Hedge currency risk through exporters and FX earners

Unless worries related to TL recede, It appears that the market is concerned about currency volatility given the sharp we would be advising companies that devaluation in the past four months. While our estimates incorporate a mild TL are hedged against TL devaluation devaluation with an USD/TL exchange rate of 3.75 by the year end, we admit that forecasting currency has become more difficult. Also, although our year-end expectation implies only 7% y/y devaluation, our average rate forecast of 3.82 USD/TL corresponds to a rather sizable devaluation of 15%. Hence, unless worries related to TL recede, we would be recommending exporters, companies with FX- linked revenues and/or international operations as the ones that have a positive operational FX leverage. Preferred names are Tupras, Enka In this context, the companies below fit well to this theme. Among these names, our Insaat, and Coca Cola Icecek preferred picks are Tupras, Enka Insaat and Coca Cola Icecek (CCI), despite the fact that the last has a short FX position.

 Commodity companies: As pricing of commodities is in hard currency, despite having operations in Turkey, Tupras, Erdemir and Petkim all generate 100% of their revenues in USD terms, whereas the share of FX in their respective cost base varies from 70% to 95%.

 Enka Insaat: Among the company’s three core operations, contracting and real estate have no exposure to Turkey. Energy is the only sector that has Turkey exposure, while this segment’s revenues are also FX-linked until the expiry of take-or-pay contracts by the end of 2018. More importantly, the company has about USD3.0bn of net cash position, corresponding to close to half of the company’s Mcap.

 Arcelik: Arcelik generates about 60% of its revenues from international markets and utilises successful hedging for its raw material procurement and balance sheet FX position. Hence, the company is rather immune to currency fluctuations.

 Kordsa: Kordsa generates all its revenues in hard currencies while 73% of its costs are in USD and EUR. Hence, the company’s operations are operationally long in FX despite the fact there is a small short FX position on its balance sheet.

 Aselsan: Although the company generates the majority of its revenues in Turkey, its contracts are predominantly in hard currency. Around 80% of the backlog is in hard currency, while 60% of costs are in FX. The company also has a strong balance sheet.

 Auto manufacturers: Both Tofas and Ford Otosan offer operational hedge, as they mostly export their productions, while the former provides better hedging with take-or-pay agreements.

Equity Research 31

 TAV Airports: TAV is operationally geared to TL weakness as it generates more than 70% of its revenues in hard currency, whereas 50% of the cost base is in FX.

 Celebi Ground Handling: Celebi generates 60% of its revenues in hard currencies while 45% of its costs are in FX. The balance sheet is also almost squared.

 Beverage companies: Both Anadolu Efes and Coca Cola Icecek generate about half of their revenues abroad. Despite both companies having short FX positions on their balance sheets, we see them as good candidates for this theme, as they show growing exposure to international markets.

 Glass companies: While we see Soda Sanayii as a dollarised business, Anadolu Cam also offers good operational hedge with its international exposure. Being the parent of these two, Sise Cam is also operationally hedged.

Theme 3: High dividend payers should outperform, at least during the dividend season

Dividends have been a working Historically, dividends have been a working investment theme for Turkish equities. investment theme for Turkish equities We looked at the share price performances of high dividend paying companies for the past eight years. We set our threshold as 5%. Assuming an investor invests in stocks that offer 5% or higher dividend yield at the beginning of the year and holds them for five months, when the dividend season ends, he/she would get a consistence outperformance. The magnitude of outperformance ranges between 4% and 28%, while the past eight years’ average outperformance stands at 12%. Moreover, 80% of stock picks have actually beat the benchmark BIST 100 index during the dividend season. Note that the stocks that we looked at are from within our coverage universe.

Figure 67: Number of stocks that offer more than 5% dividend Figure 68: Share performance (relative to BIST 100) of the yield vs number of stocks that beat BIST 100 dividend basket

18 Number of stocks with more than 5% dividend yield 30% 28% 16 Number of stock that outperformed BIST-100 during dividend seaon 25% 14 12 20% 16% 10 17 15% 13% 8 15 9% 10% 6 12 10% 8% 11 11 6% 4 9 88 8 4% 666 6 77 5% 2 5 0 0% 2009 2010 2011 2012 2013 2014 2015 2016 2009 2010 2011 2012 2013 2014 2015 2016

Source: Rasyonet, UNLU & Co estimates and analysis Source: Rasyonet, UNLU & Co estimates and analysis

Given volatility in the market, In our view, this year is unlikely to be an exception. Given volatility in the market, we investors will again be looking for believe investors will again be looking for generous dividend payers. For this year, we generous dividend payers find 12 stocks within our coverage universe that offer more than a 5% dividend yield. Some of these appear to be rather small cap and illiquid stocks, however, there a number of stocks with decent liquidity. The below table shows stocks that offer more than a 5% dividend yield.

Equity Research 32

Figure 69: Stocks that offer more than 5% dividend yield

Company DPS Share price Dividend Avg. daily trade Ticker Name (TL/share) (TL/share) yield (%) Volume (USD m) AKCNS Akcansa 1.34 14.3 9.4% 1.2 PETKM Petkim 0.37 4.08 9.2% 14.4 CIMSA Cimsa 1.48 17.36 8.5% 0.7 AYGAZ Aygaz 1.07 12.3 8.7% 0.7 TTRAK Turk Traktor 6.00 78.95 7.6% 2.3 EREGL Eregli Demir Celik 0.40 5.61 7.2% 21.1 PNSUT Pinar Sut 1.09 16.65 6.5% 0.2 FROTO Ford Otosan 1.95 32.92 5.9% 3.2 DOAS Dogus Otomotiv 0.55 9.04 6.0% 2.2 TUPRS Tupras 4.29 78 5.5% 19.1 CLEBI Celebi 1.34 25.18 5.3% 1.0 SODA Soda Sanayii 0.32 5.77 5.5% 5.2 Source: UNLU & Co estimates and analysis, Bloomberg Our best dividend picks are Tupras Although we believe that a basket of these 12 stocks, with no exclusion, should and Turk Traktor perform well, one may decide to apply a filter. We looked at the past one month relative performance to see if the market started pricing in those stocks’ dividend expectations. The chart below shows the dividend yield of each stock versus each one’s last one month performance relative to the BIST 100 performance. Accordingly, five names appear to be more attractive than others. These are Akcansa, Aygaz, Turk Traktor, Petkim and Erdemir. On the other hand, despite offering decent dividend yield, expectations on Soda Sanayii and Celebi seem to have been priced in by the market. Considering the 2017 operational outlook, liquidity and valuations, our two picks from the dividend theme are Tupras and Turk Traktor.

Figure 70: Stocks with more than 5% dividend yield and their relative performance

10%

AKCNS 9% AYGAZ PETKM

High yield High CIMSA 8% TTRAK

EREGL 7% Partially priced in

Dividend yield PNSUT DOAS 6% FROTO

TUPRS SODA 5% CLEBI

4% -4% -3% -2% -1% 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11% 12% Last one month relative performance

Source: UNLU & Co estimates and analysis, Bloomberg

Equity Research 33

Theme 4: Seeking recovery and restructuring stories

Yazicilar Holding, Coca Cola Icecek, Despite the market volatility in 2016, some companies delivered a solid recovery and Migros and Logo are our picks of this performed quite well. A similar picture might be seen in 2017 as well, in our view. theme While we have a number of candidates for this theme, we especially highlight Yazicilar Holding, Coca Cola Icecek, Migros and Logo.

 Ulker Biskuvi: Following the parent Yildiz Holding’s United Biscuits acquisition, the group is undertaking a massive restructuring and Ulker Biskuvi is taking the role of the group’s confectionery asset in Turkey, the Middle East, North Africa and Central Asia.

 Anadolu Efes: While it may not be the case at Anadolu Efes’ local beer operations, we argue that its international beer operations have seen the bottom and we anticipate an improving operating outlook, especially for Russia. In addition, the company offers a bigger restructuring story following the completion of the AB InBev and SABMiller deal.

 Coca Cola Icecek (CCI): We see CCI as a good recovery story. Over the past couple of years, the company struggled to post top-line growth in its international markets mainly due to deteriorating macroeconomic outlook, currency devaluations in Central Asian countries and the upheaval in Iraq. In our view, we saw the bottom in 2016 and expect operational recovery.

 Yazicilar Holding: On top of potential restructuring and the recovery stories of its subsidiaries Anadolu Efes and CCI, Yazicilar Holding offers a unique value unlocking story. Both majority controlling families, Yazici and Ozilhan, intend to eliminate the Anadolu Endustri Holding, which would bring a leaner holding structure. If this materialises, it would likely be a solid reason for the stock to trade at a lower holding discount.

 Migros: Following the completion of the Kipa acquisition, Migros will be facing an important restructuring in 2017. Having become successful in improving operations, we believe that management is well equipped for the restructuring.

 Logo: It is a secular growth story and we expect this will remain unchanged in the next five years. However, we look for successful restructuring and integration of the Romanian operation, which was recently added to the portfolio. Furthermore, the Romanian operation could also provide cross-selling opportunities for its core Turkey operations. Theme 5: Generating alpha through quality small caps

It has always been a good theme to invest in high quality growing off-benchmark names to generate additional alpha. We see a handful of companies that could play out well in 2017. These are TSKB, Logo and Odas Elektrik.

 TSKB: The stock remains cheap compared to its peers on a risk-adjusted basis and also relative to its own history. TSKB also remains the highest ROE generating bank within our coverage, and this is likely to continue, in our view. TSKB’s strong asset quality is another positive that differentiates the bank from its peers.

 Logo: Logo has been a strong performer in 2016 with 17% outperformance throughout the year. Yet we expect another strong performance in 2017 as well. This should be mainly on the back of the secular growth story of the company, the potential restructuring of its Romanian asset and attractive multiples.

 Odas Elektrik: The company’s 330MW lignite fired plant is scheduled to start operations in 2H17. We believe this plant will be a game changer for the company as EBITDA is expected to increase fivefold once the plant is fully up and running in 2018, leading to solid deleveraging and FCF yield going forward.

Equity Research 34

Top picks Reviewing the performance of 2016 calls

In this section, we review the performance of our picks throughout 2016. Investment themes have changed dramatically and frankly unexpectedly throughout 2016; thus, we made monthly revisions to our Top Picks with our Cherry Picks reports. Nevertheless, even if we assume no portfolio reshuffle, our Top Picks for 2016, which were stated in our Glass half full note published on 11 January 2016, have provided a positive return of 11.4% in a year, beating the benchmark BIST 100 index by 2.2pp.

Our monthly Top Picks portfolio That said, a more relevant measure of full year performance would be to account for delivered a 21% annual return in 2016, monthly portfolio reshuffles. Accordingly, our long-only ideas provided in our and significantly outperformed the monthly Cherry Picks reports delivered a 20.5% absolute return, outperforming the benchmark index benchmark BIST 100 index by 10.7pp. Throughout the year, we were able to beat the benchmark in nine months, while our performance fell below BIST 100 index in three months.

Figure 71: Performance of monthly directional long ideas

25% Directional long BIST-100 21% 20%

15% 13% 10% 10% 9% 6%6% 3% 3% 4% 4% 5% 3%2% 3% 2% 3% 0% 1% 1% 0% -2%-1% -2% -5% -3% -4% -6% -10% -8%-9%

-15% Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 2016 full year

Source: UNLU & Co estimates and analysis, Bloomberg Our pair trades posted 12.3% market In addition to providing solely directional long ideas, our Cherry Picks reports include neutral absolute return pair trade ideas as well, where we seek market neutral absolute return. The full year cumulative performance of our long/short ideas was 12.3% in 2016. While we had no pair trade ideas in the month of January, our long/short portfolio provided positive returns in nine months and negative performance in two months.

Figure 72: Performance of our pair trade ideas

120 0.4 1.2 4.3 1.0 2.0 1.0 1.3 1.4 3.9 0.2 100 3.9

80

60 112.3 100.0 40

20

0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Source: UNLU & Co estimates and analysis, Bloomberg

Equity Research 35

UNLU & Co Top Picks for 2017

We select 12 ideas based on five We have built our high conviction ideas around five investment themes described in investment themes the previous section. We select 12 stock ideas that we expect should play out well in 2017. Among banking stocks we prefer Akbank, Garanti and TSKB – also a small cap idea. Within non-financials, we like the Coke bottler Coca Cola Icecek, as we believe its international operations have seen the bottom and should recover gradually. Enka Insaat should benefit from the economic recovery in Russia and it is an obvious currency hedge with USD3.0bn net cash on its balance sheet. We like the refiner Tupras for its dollarised business and high dividends. Turk Traktor is another stock that we like because of its solid growth prospects and high dividend yield. Ulker Biskuvi is a solid international expansion story. Within retailers, we like Migro as a promising restructuring story following the acquisition of Kipa. Logo Yazilim has strong secular growth and we expect the potential restructuring of its Romanian operations to strengthen its story. We like Odas as we believe the commencement of operations of its 330MW lignite fired power plant will increase EBITDA fivefold. Lastly, Yazicilar Holding could offer a solid restructuring story once both controlling families take action to simplify the holding structure.

Among these 12 ideas, we name the most liquid ones as Top Picks for 2017. Accordingly, Akbank, Garanti, Coca-Cola Icecek, Enka Insaat, Tupras, Turk Traktor and Ulker Biskuvi are the constituents of our model portfolio.

Figure 73: Top picks for 2017

Mcap Current Target Company Name Ticker Rec. Upside Comments (TLm) Price (TL) Price (TL) Akbank has superior operational metrics such as above-sector RoE, below-sector LDR, better cost Akbank AKBNK 31,360 BUY 7.84 9.40 20% control vs peers, strong capitalisation and better CoR evolution with full NPL coverage. Garanti stands out with its above-sector RoE and capital ratios. The bank also trades at very attractive Garanti Bank GARAN 32,802 BUY 7.81 9.75 25% multiples compared to its own history and global peers. CCI has been a major underperformer since the beginning of 2014 due to the headwinds in both Coca Cola Icecek CCOLA 9,554 BUY 37.56 43.90 17% Turkey and international territories. We argue that worst is over and current valuations are attractive. Potential new contract additions and recovery in outlook for the Russian economy should be Enka Insaat ENKAI 24,444 BUY 5.82 6.92 19% supportive. Also, the company's c.USD3bn net cash makes the stock defensive in times of TL volatility. Look for 46% y/y EBITDA and 54% EPS growth in 2017. Among the prime beneficiaries of weak TL and Tupras TUPRS 19,533 BUY 78.00 91.00 17% recent natural gas tariff cut. Has best in class FCF yield, ROE & ROIC profile among blue-chips. TL depreciation and import duty tax would create headwinds for importers, which is positive for Turk Turk Traktor TTRAK 4,213 BUY 78.95 92.00 17% Traktor. Looking for 16% y/y EBITDA growth in 2017. Dividend yield (8%) is attractive. The stock trades at historical low multiples. Potentially solid 4Q16 resulta should support to reinstate Ulker Biskuvi ULKER 5,681 BUY 16.61 22.30 34% investor confidence. Long term growth story with the pladis structure remain unchanged. Source: UNLU & Co analysis

Equity Research 36

Akbank: BUY; 12M target price, TL9.40/sh

Akbank is likely to deliver strong Investment case and valuation: Akbank makes it to our preferred list owing to results in 4Q16 its superior operational metrics, including: (i) below-sector LDR; (ii) better cost control vs peers; (iii) strong capitalisation; and (iv) better cost of risk evolution with an above-sector NPL coverage ratio. Akbank delivered 65% y/y EPS growth in 9M16, and we expect the full-year EPS growth to be 51%. This should translate into a c.400bps y/y ROE (bank-only: 15.9%) improvement in 2016, according to our calculations. The management guided for 10% EPS growth and 15% ROE in 2017, on the back of flat margins, improving jaws and limited asset quality deterioration. We believe that Akbank is likely to continue to remain one of the highest ROE-generating banks within the banking sector. One of the key reasons behind our preference for Akbank is its strong capitalisation. Akbank’s bank-only CAR and Tier-I ratio stood at 15% and 13.9%, respectively, at the end of 9M16. The bank still has not revalued its fixed assets, which would add c.90bps to capital ratios, according to the bank’s management. Also, Akbank has the option to release some specific (Akbank is the only bank that currently has full NPL coverage) and general provisions in order to support both profitability and capitalisation, if needed, as it would not need as much provisioning under IFRS 9 reporting. Akbank trades at a significant discount Akbank trades at a significant discount compared to both its own history and global to its international peers and its peers, which is not justified given its improving profitability metrics. The bank historical multiples currently trades at 6.3x P/E and 0.9x P/B, based on our 2017 consolidated estimates. Catalysts: i) stronger-than-expected 4Q16 results, leading to further EPS revisions; ii) strong budget guidance for 2017 by the management; and iii) higher-than-expected loan growth with the help of the Credit Guarantee Fund. Risks: We believe that risks are very similar for all the banks in our coverage: i) CBRT remaining behind the curve and not hiking rates as expected; ii) slower-than- expected loan growth, if economic activity does not pick up and banks do not have the appetite to lend; iii) worse-than-expected asset quality deterioration, putting a pressure on profitability; and iv) further TL depreciation and a potential rating downgrade by Fitch, resulting in an erosion in capital ratios.

Figure 74: Akbank’s profitability metrics have been improving Figure 75: Akbank still trades at attractive multiples (based on one-year fwd looking P/B)

18.0% 1.3 16.2% +2 STD 16.0% 1.2 14.3% 14.6% 14.0% +1 STD 14.0% 1.1

11.9% 11.6% 12.0% 1.0

-1 STD 10.0% 0.9

8.0% 0.8 -2 STD

6.0% 0.7 2015 2016E 2017E Jul-14 Jul-15 Jul-16 Jan-15 Jan-16 Jan-17 Jan-14 Sep-14 Sep-15 Sep-16 Mar-15 Mar-16 Mar-14 May-15 May-16 May-14 Akbank Banking coverage Nov-14 Nov-15 Nov-16

Source: Company financials, UNLU & Co estimates Source: Bloomberg

Equity Research 37

Garanti: BUY; 12M target price, TL9.75/sh

We think Garanti’s 2017 budget Investment case and valuation: Garanti, once again, makes it to our preferred list guidance is optimistic and achievable owing to its strong fundamentals. The bank suffered from rising NPL inflows in 2016, and a deteriorating outlook also revealed some unexpected losses. As a result, the share performance had been relatively weak in 2016, and the bank lagged its major peers. We expect this trend to reverse in 2017. Garanti achieved 58% y/y EPS growth in 9M16, despite the additional provisioning burden from sizeable NPL files. We expect the bank to end the year with 48% y/y EPS growth, which should translate into a 350bps y/y RoE (bank-only: 15.3%) improvement. In 2017, we expect Garanti to record an EPS growth of 11% (above that of peers) and 15.0% RoE. As such, the bank is likely to remain among the highest RoE-generating banks, if not the highest, within our coverage. We highlight that the bank’s guidance of 15.5-16% RoE for 2017 poses an upside risk to our estimates. One other key factor for Garanti making into our preferred list is its strong capitalisation ratios (like Akbank). We believe that having a strong capital base will be a differentiating factor in the medium term, given the full adaptation of Basel III requirements. Garanti’s bank-only CAR and Tier-I ratio stood at 16.4% and 15.3%, respectively, at the end of 9M16. This should allow the bank to grow faster than peers if the economic outlook and loan demand improve in the future. Garanti trades at only 5.8x P/E and As is the case for the banking sector in general, Garanti trades at a significant 0.8x P/B multiples on our 2017 discount compared to both its own history and global peers. The bank currently estimates trades at 5.8x P/E and 0.8x P/B, based on our 2017 consolidated estimates. We believe that a re-rating is likely in the short term. Catalysts: i) stronger-than-expected 4Q16 results, leading to further EPS revisions; ii) strong budget guidance for 2017 by the management; and iii) higher-than-expected loan growth with the help of the Credit Guarantee Fund. Risks: The key downside risks are: i) CBRT remaining behind the curve and not hiking rates as expected; ii) slower-than-expected loan growth, if economic activity does not pick up and banks do not have the appetite to lend; iii) worse-than- expected asset quality deterioration, putting a pressure on profitability; and iv) further TL depreciation and a potential rating downgrade by Fitch resulting in an erosion in capital ratios.

Figure 76: Garanti is likely to continue to generate above-sector Figure 77: Garanti trades close to -2STD below its three-year ROE average multiples (based on one-year fwd looking P/B)

16.0% 15.4% 15.1% 1.5 15.0% 14.3% +2 STD 14.0% 1.4 14.0% 1.3 13.0% 12.5% +1 STD 1.2 12.0% 11.6% 1.1 11.0% 1.0 10.0% -1 STD 9.0% 0.9 8.0% 0.8 -2 STD 7.0% 0.7 6.0% 0.6 2015 2016E 2017E Jul-14 Jul-15 Jul-16 Jan-14 Jan-15 Jan-16 Jan-17 Sep-14 Sep-15 Sep-16 Mar-14 Mar-15 Mar-16 May-14 May-15 May-16 Garanti Banking coverage Nov-14 Nov-15 Nov-16

Source: Company financials, UNLU & Co estimates Source: Bloomberg

Equity Research 38

Coca-Cola Icecek: BUY; 12M target price, TL43.90/sh

We believe the worst is over for CCI Investment case and valuation: Coca-Coca Icecek (CCI) has underperformed the and it is the right time to reassess the BIST 100 index since the beginning of 2014 as a result of a series of unfortunate investment theme events. In our view, the term ‘headwinds’ is insufficient to describe the situation CCI has been facing, and we see the events over the past few years as more like a ‘perfect storm’. However, we believe the worst is over for CCI and it is the right time to reassess the investment theme. We have finally seen signs of The most important problems that CCI faced were the macro turmoil in the CIS improvement in the CIS operations region and security related problems in Iraq. On top of these, a temporary and Pakistan became the fastest competitive pressure in Pakistan in 2H15-1H16 further deteriorated investor growing market of CCI with a sentiment. As a result, CCI has underperformed BIST 100 by almost 40% since the growing EBITDA margin beginning of 2014. We have finally seen signs of improvement in the CIS operations following relative normalisation in the region. Competitive pressures in Pakistan, on the other hand, have already diminished and it became the fastest growing market of CCI with a growing EBITDA margin. Although problems in Iraq prevail to some extent, we believe that CCI successfully manages the situation in this region. Furthermore, CCI has already announced its appetite to acquire a controlling stake in Coca-Cola Beverages Africa (CCBA) after the Coca-Cola Company decided to exercise its call option on InBev’s controlling stake in the company. While the structure of a potential transaction is yet to be decided, we think that the likelihood of CCI becoming a much larger and more profitable bottler in the mid-term is high enough to invest in. CCI still trades at a deep discount to Despite all the supportive factors, CCI still trades at a deep discount to its historical its historical average as well as its average as well as its peers’ current average. CCI’s current one-year forward looking peers’ current average EV/EBITDA based on Bloomberg calculations is 11% lower than its mid-cycle average (Figure 79). On our numbers, CCI trades at 7.5x 2018E EV/EBITDA and 16.4x 2018E P/E, which are 13% and 10% lower than the peer average, respectively. Our DCF model also supports a potential outperformance in 2017. Catalysts: i) recovery in CCI’s international volumes; ii) normalisation in the CIS margins, especially in Kazakhstan; and iii) a potential deal involving CCBA’s controlling stake. Risks: 4Q16 financial results are likely to be weak on poor domestic margins due to an unfavourable product mix and negative bottom-line on non-cash FX and other losses. Although we do not believe a financially poor 4Q16 would affect the company’s mid-to-long-term prospects, it might still lead to a potentially short-lived weakness in the stock price. In addition, an unforeseen macro volatility in Turkey and other territories of CCI might lead to a deterioration in the consumer sentiment and delay the recovery that we have pencilled into our model.

Figure 78: y/y international volume growth reached its highest Figure 79: CCI trades below its mid-cycle average EV/EBITDA level since 3Q14 thanks to Kazakhstan and Pakistan operations multiples 30% 15 25% 14 20% 13 15% 12 10% 11 5% 0% 10 -5% 9 -10% 8 CCI's forward-looking EV/EBITDA -15% 7 3-yr average 1Q15 2Q15 3Q15 4Q15 1Q16 2Q16 3Q16 4Q16 Jul-16 Jul-15 Jul-14 Jan-17 Jan-16 Jan-15 Jan-14 Sep-16 Sep-15 International Pakistan Kazakhstan Sep-14 Mar-16 Mar-15 Mar-14 May-16 May-15 May-14 Nov-16 Nov-15 Nov-14 Source: Company data Source: Bloomberg

Equity Research 39

Enka Insaat: BUY; 12M target price, TL6.92/sh

Outlook improving with higher oil Investment case and valuation: We see Enka’s investment case as strengthening prices and stabilising rouble with gradually improving oil prices and the rouble’s strengthening relative to other EM currencies. Improving relations between Russia and Turkey and a potential easing of the sanctions against the Kremlin could only strengthen the investment case further, in our view. We foresee Enka is close to sign contracting projects in Iraq and Georgia, which potentially add close to USSD1.5bn to its contracting backlog, which stood at USD1.6bn as of 3Q16. While the company’s appetite seems to be lower to take part in EPC contracts related to the Indonesian government’s 35GW power capacity expansion projects, it may well be a bidder as an investor to become an IPP in the country. Rental and occupancy rates in the Russian real estate market, for both the office and retail segments, have declined sharply in the past two years. With the improving outlook of the Russian economy, we see more risks on the upside than downside, and 2Q16 and 3Q16 results support our view. Enka’s favourable take-or-pay contract will terminate by the end of 2018 and it will be selling electricity on the free market starting from 2019. Inevitably, profitability levels will decline but we still see solid EBITDA generation from Enka’s three power plants.

Management is scrutinising new While Enka’s sizeable cash pile (33% of NAV in hard currency) makes it defensive opportunities for better utilisation of during times of currency volatility, we believe that the company is eyeing a number of cash potential investments to better utilise its cash. Given the company’s conservative approach in evaluating new projects, we see a high likelihood that entering a new business will be value accretive. Catalysts: (i) strengthening rouble; (ii) rising oil prices; and (iii) adding a new contracting project — power projects in Iraq and Georgia appear solid candidates. Risks: (i) Delayed recovery in the Russian economy; (ii) a failure to get sizeable new projects; (iii) a steeper-than-anticipated fall-off in electricity sales after the termination of the BOT contract; and (iv) potential financial loss to be booked in 4Q16 on the back of market volatility and Enka’s cash position of over USD3.0bn.

Figure 80: Enka Insaat – EBITDA forecasts (USDm) Figure 81: Enka Insaat’s NAV breakdown (%)

Contracting 1,000 Contracting Real estate Energy Other 14% 900 876 800 763 742 706 Cash 700 666 648 33% 600 500 400 300 Real estate 33% 200 Other 100 2% 0 Energy 2014 2015 2016E 2017E 2018E 2019E 18% Source: Company data, UNLU & Co analysis and estimates Source: UNLU & Co analysis and estimates

Equity Research 40

Tupras: BUY; 12M target price, TL91.00/sh

Investment case and valuation: According to OPEC, total distillation capacity additions for 2016-2021 are projected at 7.35m bbl/day, suggesting further overcapacity in the market. The caveat is that 50% of the new projects are far from the construction stage. We believe Tupras is well positioned as: i) it improved its position on the supply curve through RUP; ii) the company has a low cost operation owing to weak TL and a lean organisational structure; iii) Tupras has a great feedstock flexibility, which is crucial in an environment where supply reliability is overrated; and iv) the company has secure access to end markets as the sole refinery in Turkey until 2019, and has a retail arm (OPET) that ensures sales/distribution of the products. Tupras’ profitability remained under severe pressure in 1H16 due to two major headwinds: i) extraordinary weakness in middle distillate cracks; and ii) the fire accident that led to a production halt in the hydrocracker unit at RUP for two months (3 Feb-7 Apr). Hence, Tupras’ premium over Med Complex declined to its lowest level in 1Q16 and remained way below its mid-cycle average in 2Q16. However, we see initial signs of mean reversion in 3Q16, which is encouraging for the upcoming quarters, in our view. We expect Tupras to deliver strong We expect Tupras to deliver strong earnings growth in 2017 on the back of: i) a earnings growth in 2017E higher premium over Med Complex, stemming from a recovery in middle distillate crack margins; ii) lower natural gas tariffs; and iii) depreciation in TL. Our 2017/18 EBITDA and net income forecasts are 10% ahead of consensus. Currency devaluation helps Tupras to Following the RUP, Tupras’ annual natural gas consumption climbed to c.1.3bcm. improve its operational profitability Incorporating both the recent TL depreciation and natural gas tariff cut, we expect Tupras to save USD14m in 4Q16 vs 3Q16. Total savings will reach USD70m in 2017 vs 2016. OPEX is another area where Tupras has created a significant advantage owing to: i) weak TL; and ii) rising CUR. As such, OPEX/bbl has come down by 43% since 2010 Catalysts: i) upcoming quarterly results (1Q and 2Q in particular); ii) normalisation in mid-distillate cracks; and iii) a higher-than-expected dividend announcement. Risks: i) lower-than-projected refining trends/margins; ii) significant volatility in crude oil prices; iii) SOCAR’s refinery commencing operations earlier than expected (base case: 2019); and iv) a potential fine by the competition board or the tax authority

Figure 82: We remain conservative on our long-term forecasts, Figure 83: Tupras trades below its mid-cycle average on one-year which are below mid-cycle operating profitability FW EV/EBITDA multiple

USD/bbl Net Operating Profit 12.0 8.0 10-yr avg. Net Operating Profit (USD3.9/bbl) Tupras 7-yr avg. 11.0 7.0 10.0 6.0 5.0 9.0 4.0 8.0 3.0 7.0 2.0 6.0 1.0 5.0 0.0 4.0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Jul-16 2016E 2017E 2018E 2019E 2020E 2021E Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16

Source: Company data Source: Bloomberg

Equity Research 41

Turk Traktor: BUY; 12M target price, TL92.00/sh

Investment case and valuation: 2017 is likely to be a challenging year for all Turkish discretionary sectors as the deterioration in consumers’ affordability will weigh negatively on demand. That said, the Turkish tractor sector is much more stable compared to Turkish autos and parts, thanks to continued support from the government for the agriculture sector. This is the main reason why we expect only a 3% y/y contraction in the Turkish tractor market in 2017 vs a 9% y/y contraction in the Turkish auto sector. We expect Turk Traktor to gain a More importantly, we see two main competitive advantages of Turk Traktor in 2017 competitive edge in 2017 with TL that might lead to market share gain: i) TL depreciation is a big headwind for depreciation and the recently importers that gained notable market share in the past two years; and ii) the recently implemented import duty tax implemented import duty tax (+21%) for some countries would increase importers’ cost further. In light of these developments, we expect Turk Traktor to gain 2.0% market share in 2017. This will lead to +1% y/y domestic volume growth. Moreover, the import duty tax could also ease competition and hence provide further room for the market leader to pass through cost inflation to consumers. The dividend yield should support On the exports, we expect a normalisation in the recent downtrend and look for flat share price performance in 1H17 volumes in 2017. We note that in the past two years, Turk Traktor’s exports volume declined on average by 11% due to sluggish global tractor demand. In aggregate, we expect Turk Traktor to increase its volumes by 1% y/y in 2017. We expect profitability to remain strong in 2017 and look for 16% y/y EBITDA growth. Last but not least, the high dividend yield (7.7%) is another catalyst for the share price performance ahead of dividend season. Besides the tractor business, Turk Traktor also started to import construction equipment in late 2014. The Turkish construction equipment sector is one of the largest in the world with a USD6.5bn turnover. Although the competition is very fierce in the sector, the local producers only meet 40% of the demand in Turkey. As Turk Traktor is a new player in the sector, the company is trying to gain market share and build local customers in this segment. In the long run, we think Turk Traktor might invest in this segment and became a manufacturer of light construction equipment in its Erenler plant as the production process of some construction equipment units is very similar to tractor production. If it materialises, this will be positive for the company, with diversification in its portfolio and opportunities for a stronger growth. Catalysts: (i) a recovery in monthly tractor sales; (ii) potentially strong 4Q16 results; and (iii) high dividend yield. Risks: (i) a sharp decline in loans provided by Ziraat Bank; (ii) continued tough competition from importers; and (iii) a further decline in exports.

Figure 84: Market share of Turk Traktor and importers in the Figure 85: Turk Traktor one-year forward looking EV/EBITDA domestic market multiple

NH + CIH Other local players Imported tractors One-year fwd EV/EBITDA average -1 STD +1 STD 100% 12.0 +2 STD -2 STD 21% 23% 21% 22% 24% 27% 29% 80% 11.0

25% 29% 27% 31% 28% 10.0 60% 25% 25% 9.0 40% 8.0

52% 50% 50% 48% 49% 48% 20% 46% 7.0

6.0 0% 2010 2011 2012 2013 2014 2015 11M16 Jul-16 Jul-15 Jul-14 Jul-13 Jan-17 Jan-16 Jan-15 Jan-14 Jan-13 Sep-16 Sep-15 Sep-14 Sep-13 Mar-16 Mar-15 Mar-14 Mar-13 May-16 May-15 May-14 May-13 Nov-16 Nov-15 Nov-14 Nov-13 Source: Company data, UNLU & Co analysis and estimates Source: Bloomberg

Equity Research 42

Ulker Biskuvi: BUY; 12M target price, TL22.30/sh

Ulker Biskuvi is set to consolidate the Investment case and valuation: Ulker shares have tumbled by 24%, group’s confectionery operations in underperforming the BIST 100 index by 29% since 4 October, 2016. This was mainly Turkey, the Middle East, North Africa due to allegations that appeared in some unrespectable websites and its 3Q16 and Central Asia financials, which were perceived as weak. We believe the sales volume decline was a one-off and the company’s EBITDA generation was solid, despite soft volumes. The Turkish consumer sector experienced a channel-related abnormality following the failed coup attempt. People refrained from going out for three days (impacting volumes by 3%, i.e. three days in 92 days), 10+ retailers went bankrupt and Turkey’s largest retailer A101 stopped its procurement process for a few days. In all, 3Q16 was an exception, hence the volume weakness was a one-off, in our view. Following parent Yildiz Holding’s United Biscuits acquisition, the group has been restructuring all of its confectionery businesses. The strategy is to manage all brands and products together, which should create significant synergies for the group. The group has also decided to manage all operations regionally and Ulker Biskuvi is set to consolidate the group’s confectionery operations in Turkey, the Middle East, North Africa and Central Asia. Having operations in Turkey and already exporting to most of the countries in these regions, the group is likely to capitalise on Ulker’s know- how. Ulker Biskuvi is also likely to benefit from economies of scale, a wider product portfolio and diversified geographies.

We see significant room for Ulker We see significant room for Ulker Biskuvi to leverage parent Pladis’s structure. This Biskuvi to leverage parent Pladis’s should come in the form of accessing a wider range of products, R&D and structure procurement synergies, and logistics advantages. Assets under Pladis, namely United Biscuits, Godiva and DeMet’s, have a wide-ranging product offering. Ulker Biskuvi has the opportunity to launch any of these to further penetrate the markets that it operates in. A few product launches have already taken place. The group could centralise its R&D efforts, which should not only lower costs but also improve the number of winning new product launches. Furthermore, centralised procurement for all Pladis companies should improve bargaining power, leading to better purchasing terms. Last but not least, having wider operating geographies will provide logistics advantages, as the company was exporting solely from Turkey but exporting from a closer country could provide cost advantages. Catalysts: (i) further acquisitions at favourable valuations to strengthen Ulker’s presence in the region; (ii) growing product portfolio with more value-added products within the Pladis structure; and (iii) improving consumer sentiment. Risks: (i) weakening TL, which could translate into cost pressure; (iii) deterioration in consumer confidence; and (iii) a rise in soft commodity prices (cocoa and palm oil).

Figure 86: One-year forward EV/EBITDA is at historical lows Figure 87: Ulker’s revenue (TLm) and EBITDA margin forecasts

17 6,000 13.6% 14.0% 16% +2 STD 13.3% 13.1% 16 14% 5,000 11.5% 15 12% 4,000 14 10% 13 3,000 8% 3-year historical

12 5,075 6% avg EV/EBITDA 2,000 4,482

11 3,957 4% 3,075 1,000 2,891 10 -2 STD 2% 9 0 0% 8 2014 2015 2016E 2017E 2018E

Revenues EBITDA margin (rhs) Jul-16 Jul-15 Jul-14 Jan-16 Jan-15 Jan-14 Sep-16 Sep-15 Sep-14 Mar-16 Mar-15 Mar-14 May-16 May-15 May-14 Nov-16 Nov-15 Nov-14 Source: Bloomberg Source: Company data, UNLU & CO estimates

Equity Research 43

Disclosure Appendix

Important Global Disclosures

The information and opinions in this research report was prepared by UNLU Menkul Degerler A.S ("UNLU & Co").

For important disclosures, stock price charts and equity rating histories regarding companies that are the subject of this report, please contact UNLU & Co Research and / or Compliance - +90 212 367 3636.

For valuation methodology and risks associated with any price targets referenced in this research report, please email: [email protected] with a request for valuation methodology and risks on a particular stock.

The following analyst/s: Vedat Mizrahi, Can Ozguzel, Can Oztoprak, Emir Moran, Esra Simsek, Mete Ozbek, Muharrem Gulsever, Sinan Veziroglu certify(ies), with respect to the companies or securities under analysis, that (1) the views expressed in this report accurately reflect his/her/their personal views about all of the subject companies and securities and (2) no part of their compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report.

* Any other material conflict of interest of the research analyst or member (Unlu Securities Inc., the US broker-dealer) that the research analyst or an associated person of the member with the ability to influence the content of a research report knows or has reason to know at the time of the publication or distribution of a research report is as follows: NONE

See the Companies Mentioned in Figure 46 on Page 23 for full company names.

Analysts’ stock ratings are defined as follows*: Buy (B): The stock’s total return* is expected to be more than 20% (or more, depending on perceived risk) over the next 12 months. Hold (H): The stock’s total return is expected to be in the range of 10-20% over the next 12 months. Sell (S): The stock’s total return is expected to be less than 10% over the next 12 months. Restricted (R): In certain circumstances, UNLU & Co and/or applicable law and regulations preclude certain types of communications, including an investment recommendation, during the course of UNLU & Co’s engagement in an investment banking transaction and in certain other circumstances. Speculative Buy: UNLU & Co may issue a “Speculative Buy” when the Research Analyst covering the Company is of the view that the risk/reward tradeoff is somewhat less compelling than that of a BUY rating. These companies tend to have very high upside potential, but also a great degree of risk or uncertainty with regard to future financial results. Relative Three Month Ratings: UNLU & Co may also assign a three-month relative call (or rating) to a stock to highlight expected out-performance (most preferred) or under-performance (least preferred) versus the geographic and industry sector over a three (3) month period. The relative call may highlight a specific near-term catalyst or event impacting the Company or the market that is anticipated to have a short term price impact on the equity securities of the Company. Absent any specific catalyst the analyst(s) will indicate the most and least preferred stocks in the universe of stocks under coverage, explaining the basis for this short term view. This three month view may be different from and does not affect a stocks` fundamental equity rating, which reflects a longer-term total absolute return expectation.

Volatility Indicator [V]: A stock is defined as volatile if the stock price has moved up or down by 20% or more in a month in at least 8 of the past 24 months or the analyst expects significant volatility going forward.

Analysts’ coverage universe weightings are distinct from analysts’ stock ratings and are based on the expected performance of an analyst’s coverage universe* versus the relevant broad market benchmark**: Overweight: Industry expected to outperform the relevant broad market benchmark over the next 12 months. Market Weight: Industry expected to perform in-line with the relevant broad market benchmark over the next 12 months . Underweight: Industry expected to underperform the relevant broad market benchmark over the next 12 months. Not Covered: UNLU & Co Equities Research does not cover the issuer or offer an investment view on the issuer or any securities related to it. Any communication from Research on securities or companies that UNLU & Co does not cover is factual or a reasonable, non-material deduction based on an analysis of publicly available information or consensus forecasts

*Total return is calculated as the sum of the stock’s expected Capital Appreciation and expected Dividend Yield.

*UNLU & Co Small and Mid-Cap Advisor stock: Stock ratings are relative to the Borsa Istanbul A.S ("BIST") index.

**An analyst's coverage universe consists of all companies covered by the analyst within the relevant sector.

Equity Research 44

UNLU & Co distribution of stock ratings is:

Ratings Distribution as of the date of this report BUY Hold Sell RESTRICTED All Recommendations (%) 52 45 3 0

* our stock ratings of BUY, HOLD, and SELL most closely correspond to Buy, Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's decision to buy or sell a security should be based on investment objectives, current holdings, and other individual factors.

UNLU & Co policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the market that may have a material impact on the research views or opinions stated herein.

UNLU & Co policy is only to publish investment research that is impartial, independent, clear, fair and not misleading. For more detail please contact the Compliance Division of UNLU & Co and request their Policies for Managing Conflicts of Interest in connection with Investment Research.

UNLU & Co does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot be used, by any taxpayer for the purposes of avoiding any penalties.

Unlu Menkul Degerler A.S. (“UNLU & Co”), is authorized and regulated by the Capital Markets Board of Turkey (“CMB”) and a member of Borsa Istanbul A.S. (“BIST”). Under CMB’s legislation, the information, comments and recommendations contained in this report fall outside of the definition of investment advisory services. Investment advisory services are provided by authorized entities by taking into account the risk and return preferences of the concerned persons. The comments and recommendations contained in this report have general nature. These recommendations may not fit to your financial status, risk and return preferences. For this reason, to make an investment decision by relying solely to this information stated here may not bring about outcomes that fit your expectations

Company Specific Disclosures: Important Disclosures, including price charts are available for compendium reports and all UNLU & Co covered companies by emailing [email protected] or calling +90 212 367 3690 with your request. UNLU & Co Research team may screen companies not covered by UNLU & Co. For important disclosures for these companies, please call + 90 212 367 3817 or e- mail [email protected]

Important Regional Disclosures

This report covers Turkish Equities. All other companies were used for illustrative purposes only. We are not commenting on the investment merit of the securities of these companies Singapore recipients should contact a Singapore financial adviser for any matters arising from this research report. The analyst(s) involved in the preparation of this report have not visited the material operations of the subject company(ies) within the past 12 months. As of the date of this report, UNLU & Co does not act as a market maker or liquidity provider in the equities securities that are the subject of this report. Principal is not guaranteed in the case of equities because equity prices are variable. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. Investors should carefully consider their own investment risk. Investment results are the responsibility of the individual investor. Reports may not be reprinted without permission of UNLU & Co.

Equity Research 45

Important UNLU & Co Disclosures

Potential Conflicts Company Disclosure Anadolu Cam D & F Aksa Enerji D Akfen D & G Ayen Enerji D & F Çalik Enerji D & F IC Ictas Enerji D Karsan Otomotiv D Gozde Girisim D & G Palmet Enerji D & G Soktas Tekstil D Ulker Biskuvi D. F & G

A: The analyst, a team member, a member of the analyst's household or a team member's household serves as an officer, director or advisory board member of the subject company B: The company beneficially owns 5% or more of the equity shares of UNLU & Co as at date of this report C: UNLU & Co beneficially owns 1% or more of the equity shares of the company D: The Company is a client of UNLU & Co E: UNLU & Co has lead managed or co-lead managed a public offering of securities in the Company or any related derivatives in the last 12 months F: UNLU & Co has received compensation for investment banking services from the company within the last 12 months G: UNLU & Co expects to receive, or intends to seek, compensation for investment banking services from the company during the next 3 months H: UNLU & Co has sent extracts of this research report to the subject company prior to publication for the purpose of verifying factual accuracy. Based on information provided by the subject company, factual changes have been made as a result. I: Analyst or a member of their household holds long or short personal positions in a class of common equity securities of this company J: UNLU & Co is a market maker or liquidity provider in the financial instruments of the relevant issuer or any related derivatives K: UNLU & Co provided non-investment banking services, which may include Sales and Trading services, to the subject company within the past 12 months L: UNLU & Co has received compensation for products and services other then investment banking services from the subject company within the past 12 months M: UNLU & Co beneficially owns 5% or more of the equity shares of the Company

* Disclosures are correct as of date of this report

For purposes CMB, in connection to the distribution of UNLU & Co research, UNLU & Co must disclose certain material conflicts of interest.

This report may include references to UNLU & Co’s research recommendations. For further information and for UNLU & Co reports in their entirety, please visit the website at www.unlumenkul.com

For UNLU & Co disclosure information on other companies mentioned in this report, please visit the website at www.unlumenkul.com.

Disclaimers continue on next page.

Equity Research 46

Disclaimer and Confidentiality Note

This report is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject UNLU & Co to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to UNLU & Co. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of UNLU & Co. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of UNLU & Co or their subsidiaries.

The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. UNLU & Co may not has taken any steps to ensure that the securities referred to in this report are suitable for any particular investor. UNLU & Co will not treat recipients as their customers by virtue of their receiving the report. The investments or services contained or referred to in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report constitutes investment, legal, accounting or tax advice or a representation that any investment or strategy is suitable or appropriate to your individual circumstances or otherwise constitutes a personal recommendation to you. UNLU & Co does not offer advice on the tax consequences of investment and you are advised to contact an independent tax adviser. Please note in particular that the bases and levels of taxation may change.

UNLU & Co believe the information and opinions in the Disclosure Appendix of this report are accurate and complete. Information and opinions presented in the other sections of the report were obtained or derived from sources UNLU & Co believe are reliable, but UNLU & Co makes no representations as to their accuracy or completeness. Additional information is available upon request. UNLU & Co accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that liability arises under specific statutes or regulations applicable to UNLU & Co. This report is not to be relied upon in substitution for the exercise of independent judgment. UNLU & Co may have issued, and may in the future issue, a trading call regarding this security. In addition, UNLU & Co may have issued, and may in the future issue, other reports that are inconsistent with, and reach different conclusions from, the information presented in this report. Those reports reflect the different assumptions, views and analytical methods of the analysts who prepared them and UNLU & Co is under no obligation to ensure that such other reports are brought to the attention of any recipient of this report. UNLU & Co are involved in many businesses that relate to companies mentioned in this report.

Descriptions of any company or issuer or their securities or the markets or developments mentioned in the Research are not intended to be complete. The Research should not be regarded by recipients as a substitute for the exercise of their own judgment as the Research has no regard to the specific investment objectives, financial situation or particular needs of any specific recipient.

Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication by UNLU & Co and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADRs, the values of which are influenced by currency volatility, effectively assume this risk.

Structured securities are complex instruments, typically involve a high degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (including, but not limited to, spot and forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a structured product should conduct their own investigation and analysis of the product and consult with their own professional advisers as to the risks involved in making such a purchase.

Some investments may not be readily realizable since the market in the securities is illiquid or there is no secondary market for the investor’s interest and therefore valuing the investment and identifying the risk to which the investor is exposed may be difficult to quantify. Investments in illiquid securities involve a high degree of risk and are suitable only for sophisticated investors who can tolerate such risk and do not require an investment easily and quickly converted into cash. Other risk factors affecting the price, value or income of an investment include but are not necessarily limited to political risks, economic risks, credit risks and market risks.

Some investments discussed in this report have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment, in such circumstances you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed.

Equity Research 47

UNLU & Co maintains information barriers between their Research Analysts and the rest of their and their shareholders business divisions, more specifically the Investment Banking business. UNLU & Co analysts’, strategists’ and economists’ compensation is not linked to Investment Banking or Capital Markets transactions performed by UNLU & Co or their shareholders. Facts and views presented in UNLU & Co research has not been reviewed by, and may not reflect information known to, professionals in other UNLU & Co business areas, including investment banking personnel.

This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of UNLU & Co has not reviewed the linked site and takes no responsibility for the content contained therein. Such address or hyperlink (including addresses or hyperlinks to UNLU & Co’s own website material) is provided solely for your convenience and information and the content of the linked site does not in any way form part of this document. Accessing such website or following such link through this report shall be at your own risk.

In jurisdictions where UNLU & Co is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. US investors transacting in the securities featured or mentioned in this research report must deal directly through a U.S. Registered broker-dealer.

This document does not constitute or form part of, and should not be construed as, an offer or invitation to subscribe for or purchase any securities, and neither this document nor anything contained herein shall form the basis of or be relied on in connection with or act as an inducement to enter into any contract or commitment whatsoever. This document has not been published generally and has only been made available to institutional investors. Any decision to subscribe for or purchase securities in any offering must be made solely on the basis of the information contained in an offering memorandum (and supplements thereto) or any other offering document issued in connection with any proposed offering.

UNLU & Co does not form a fiduciary relationship or constitute advice and this Research is not and should not be construed as an offer or a solicitation of an offer of securities or related financial instruments or an invitation or inducement to engage in investment activity, and cannot be relied upon as a representation that any particular transaction necessarily could have been or can be effected at the stated price.

Please note that this report was originally prepared by UNLU & Co for distribution to market professionals and institutional investor customers. Recipients who are not market professionals or institutional investor customers of UNLU & Co should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents.

UNLU & Co is a member of the BIST

Copyright 2017 UNLU & Co All rights reserved.

.

Equity Research 48

Head of Equity & Credit Research Head of Equity Sales Head of Equity Trading Vedat Mizrahi, PhD (90 212) 367 3690 Tunc Yildirim (90 212) 367 3675 Batur Ozyar (90 212) 367 3673 [email protected] [email protected] [email protected]

Equity Research Team Equity Sales Team Equity Trading Team Banks & Insurance Turkey Turkey Vedat Mizrahi, PhD (90 212) 367 3690 Tunc Yildirim (90 212) 367 3675 Batur Ozyar (90 212) 367 3673 [email protected] [email protected] [email protected] Emir Moran (90 212) 367 3687 Kagan Cevik (90 212) 367 3683 Engin Cebeci (90 212) 367 3674 [email protected] [email protected] [email protected] Real Estate Yasemin Ahmetoglu (90 212) 367 3671 Emir Moran (90 212) 367 3687 [email protected] [email protected] Ozlem Turgay (90 212) 367 3676 Airlines [email protected] Muharrem Gulsever (90 212) 367 3694 [email protected] United States Autos & Parts Batur Ozyar (90 212) 367 3673 Can Ozguzel (90 212) 367 3678 [email protected] [email protected] Retail & Beverage Mete Özbek (90 212) 367 3689 [email protected] Food Can Oztoprak (90 212) 367 3692 [email protected] TMT Can Oztoprak (90 212) 367 3692 [email protected] White Goods Mete Özbek (90 212) 367 3689 [email protected] Conglomerates Can Oztoprak (90 212) 367 3692 [email protected] Oil and Gas Muharrem Gulsever (90 212) 367 3694 [email protected] Metals and Mining Muharrem Gulsever (90 212) 367 3694 [email protected] Building Materials Can Ozguzel (90 212) 367 3678 [email protected] Economist Esra Șimșek (90 212) 367 3696 [email protected]

Credit Research Team Credit Sales Team Sinan Veziroglu (90 212) 367 3691 Tunc Yildirim (90 212) 367 3675 [email protected] [email protected] Vedat Mizrahi, PhD (90 212) 367 3690 Yasemin Ahmetoglu (90 212) 367 3671 [email protected] [email protected] Ozlem Turgay (90 212) 367 3676 [email protected]

Equity Research 49