We turn selective after the recent rally. We maintain our view that profitability should improve significantly in the next two years as provision expenses normalize from their very high base. That said, recent rate hikes and lockdowns have toughened the near-term outlook and we now expect the earnings recovery to commence in 2H21e rather than 1H21e. This delay, and the 40% rally in the last 3
months, call for a selective approach when playing the earnings recovery theme. We trim FY21e earnings 7% and raise FY22e 4% to bring estimates in line with this outlook, and adjust TPs. We reiterate Buy on Akbank and Garanti due to better earnings momentum and wider valuation discounts, but downgrade YKB following its significant rerating. We now have only two banks as Buys versus all four in Nov’20. Outlook for FY21e.
Margins to contract sharply in 1H21e as CBRT’s rate hikes to stabilize the currency and tame inflation elevate funding costs. Come 2H21e, however, we expect inflation to moderate, funding costs to ease and NIMs to recover. As for asset quality, we expect 3.5% NPL formation in FY21e after no inflows last year thanks to moratoria. Yet, banks’ 300bps CoR in FY20e suggests that most of this year’s flow has been pre-provisioned and hence we expect CoR to moderate to 225bps in FY21e and to 175bps in FY22e. This improvement will be masked in 1H21e by margin weakness and normalizing trading income, but should become visible thereafter. On a full year basis we expect earnings to grow 8% in FY21e, but
jump 65% in FY22e owing to the low base of 1H21e. Reiterate Buy on Akbank, Garanti; downgrade YKB to Hold. In the absence of last year’s FX losses on the Turk Telekom loan and margin pressures from asset
ratio regulation, FY21e earnings momentum should be better at Akbank and Garanti vs. peers. As such, their valuations still look discounted vs. history and EEMEA banks, that trade at >1.0x P/B for similar ROTE and capital profiles. We downgrade YKB to Hold; its comps have reached those of Akbank, which looks stretched given the latter’s stronger capital. We maintain Hold on Isbank.
Akbank (AKBNK TI, Buy TP TRY8.15 (from TRY6.75) – We expect the transitory headwinds that depressed earnings in FY20 and caused the shares to underperform peers to subside this. As such, we see profitability (ROE) climbing back towards 14.5% in 2H21e once inflation and funding costs ease. With such an ROE Akbank has traded at around 0.9x P/B in the past vs. the current 0.6x. Following its underperformance, Akbank’s valuation is on par with that of YKB which looks unwarranted to us considering the former’s higher capital-adjusted ROE.
Garanti (GARAN TI, Buy, TP TRY12.10 (from TRY11.90) – We fine-tune our estimates for the higher than expected rate hike at the last MPC and reiterate our bullish view that we recently published in Garanti: Strong ROE recovery after challenging first half, 18 December 2020. We expect Garanti’s ROE to improve towards 16% in 2H21e, after a challenging start to the year, as its NIM recovers from the pressure of high funding costs and its provision expenses decrease from their very high base. With a similar ROE the stock has traded between 0.8x and 1.0x trailing P/B in the past versus the current 0.7x.
Isbank (ISCTR TI, Hold, TP TRY8.00 (from TRY7.55) – We see a balanced risk reward after a 31% rise in the share price in the last three months. We expect Isbank to face somewhat higher pressure than peers at the PPI level this year as it operates with a higher cost base, which will increase due to inflation, whereas its revenues will be pressured by higher funding costs. We consider the stock’s 13% discount to its peers on P/PPI as a fair reflection of this situation.
YKB (YKBNK TI, Downgrade to Hold, TP TRY3.35 (from TRY2.60) – YKB’s significant rerating in the past 12M has brought its 1Y forward PE of 5.6x and trailing P/B of 0.6x almost at par with Akbank and Garanti. Though we consider the rerating as fair, we also struggle to see why YKB should start trading at a premium valuation. First, the ROE conversion between YKB and peers in FY20 was partly because the latter had non-recurring currency losses on their Turk Telekom loan. Second, for YKB’s ROE to sustainably exceed its peers, its PPI margin has to converge with theirs. But, the last two years’ performance don’t give us a strong conviction on that front. We downgrade to Hold due to the unenticing relative value proposition and the challenging near-term earnings outlook.
The recent changes in economic management in Turkey, and the more orthodox monetary policy
framework communicated to the market, lead us to forecast a V-shaped pattern for NIMs in the next
12 months. Following CBRT’s rate hikes, high-end TRY deposit costs have reached almost 17%. We
expect them to stay at current levels in 1H21e and then start to decrease for the following reasons:
1. TRY has been deeply undervalued in REER terms for a while and foreign ownership of local
currency bonds hit historical lows this year. However, the missing ingredient for a positive
backdrop was positive real rates, which finally arrived with CBRT’s rate hikes after September.
The ex-post real rate is currently 250bps, but based on HSBC Economics’ 11.6% YE21e
inflation forecast the real rate buffer would be 540bps ex-ante. Considering that many other EM
countries are offering negative real rates, the carry looks appealing in Turkey
2. On average TRY deposits are yielding a 4% ex-ante and post-tax real return. While this
may not be enough to trigger de-dollarization of savings, it could limit the increase in
marginal dollar demand by retail investors and thereby help with currency stabilization.
3. With stability in the currency, our economists expect inflation to moderate from 14.6%
currently towards 11.6% by YE21e. This improvement could give Turkish banks some room to lower their deposit costs in 2H21e The reflection of all this on margins is depicted in Figure 18. Given TR banks’ duration mismatches, we expect NIM to contract sharply between 3Q20 and 2Q21e. Come 2H21e
however, we expect inflation to moderate funding costs to ease and NIM to recover.
We expect the PPI pattern to mimic that of net interest margins, as shown in Figure 18: a sharp
contraction in 1H21e followed by a recovery. Weakness in net interest income in 1H21e and the
absence of last year’s robust trading gains could lead to a 14% contraction in PPI in FY21e. The
projected recovery in NIMs, and the low base of 1H21e, lead to a forecast 34% expansion for FY22e.
Source: HSBC Global Research