INFLATION-LINKED BONDS: NEW ASSET CLASS FROM MINFIN

Transcription

1 INFLATION-LINKED BONDS: NEW ASSET CLASS FROM MINFIN Before end July the MinFin could place a maiden issue of Canadian model inflation-linked bonds (OFZ-IN). The placement will write a new chapter in the history of the Russian financial market, marking the appearance of a new asset class, which would allow to hedge inflation risks and diversify investment in Russian assets. Moreover, the new paper will provide investors with a more accurate market estimate of inflation expectations. Key investors in this instrument should be Russian institutional investors, primarily pension funds, as well as foreign investors on expectations of inclusion of OFZ-IN into benchmark inflation-linked indexes. Medium-term inflation forecasts currently envision a wide range of figures from 5-10%. RUB corporate bonds with coupons linked to CPI are currently pricing inflation of ~ %. Based on market-implied inflation expectations, we estimate the real yield on the 8Y OFZ-IN in the range of ~ %. New chapter in the Russian financial market s evolution Inflation-linked bonds (ILB) are a well-known instrument in both developed and emerging debt markets. In the US and UK (the two largest ILB markets), ILB accounts for 8% and 19% of sovereign debt, respectively, while in emerging markets the share of such bonds ranges from 12% to 35%.. Alexey Demkin, CFA +7 (495) Alexey.Demkin@gazprombank.ru Yakov Yakovlev +7 (495) Yakov.Yakovlev@gazprombank.ru Timur Semenov +7 (495) (ext ) Timur.Semenov@gazprombank.ru Artem Beketov +7 (495) (ext ) Artem.Beketov@gazprombank.ru Table 1. Volumes of inflation-linked bonds in curculation in DM and EM, $ bln TURKEY BRAZIL MEXICO SOUTH AFRICA US UK Local sovereign debt market ,906 2,252 Inflation-linked bonds , Share 22% 35% 12% 16% 8% 19% The main advantages of ILB for investors are: Source: Bloomberg hedging investment from inflation and the opportunity to lock in a long-term positive real return; investment diversification: the real yield of inflation-linked papers historically shows weak correlation with other asset classes (traditional bonds, equities); lower volatility compared with other assets. The main advantages for sovereign borrowers: They improve the credibility of a country s monetary policy, especially if it is one that targets inflation. By placing inflation-linked papers, a borrower exhibits confidence that inflation will decline in the future. Expansion of the sovereign debt investor base. In developed and EM markets, inflation-linked bonds have already formed a separate asset class. A potential decrease in borrowing costs. By placing inflation-linked bonds, an issuer does not have to pay an inflation risk premium, which is contained in a conventional bond s yield. 1 Copyright

2 JAN 12 MAY 12 SEP 12 JAN 13 MAY 13 SEP 13 JAN 14 MAY 14 SEP 14 JAN 15 MAY 15 JAN 04 SEP 04 MAY 05 JAN 06 SEP 06 MAY 07 JAN 08 SEP 08 MAY 09 JAN 10 SEP 10 MAY 11 JAN 12 SEP 12 MAY 13 JAN 14 SEP 14 MAY 15 JULY 10, 2015 Chart 1. Total Return for different asset classes for the past 10 years (RUB terms), cumulative inflation and modeled OFZ-IN bond JAN 1, 2004= RUBLE BONDS* EUROBONDS* MICEX INDEX** GOLD COMMODITIES MARKETS*** DUAL-CURRENCY BASKET INFLATION MODELED OFZ-IN BOND**** Source: Bloomberg, Gazprombank estimates * indexes for Russian bonds denominated in rubles and dollars are calculated by Gazprombank ** does not include the dividend yield *** CRB Spot All Commodities Index **** modeled inflation-linked OFZ issue. Calculations are based on the historical dynamic of OFZ yields, indexation to inflation is in line with the terms of planned OFZ-IN. The expected inflation is taken from the Economy Ministry forecasts. Inflation significantly erodes the long-term return on investment in ruble-denominated bonds. According to our estimates, since January 1, 2012 the cumulative Total Return on modeled inflation-linked bonds (in accordance with formulas given in terms of issuance of OFZ-IN), would have amounted to 45%, or three times the Total Return on traditional OFZs (about 16% for the long-term OFZ 26205). Chart 2. Total Return for modeled 8Y inflation-linked OFZ and OFZ JAN 1, 2012 = MARKET ESTIMATE OF INFLATION EXPECTATIONS IN OFZ Source: Bloomberg, Gazprombank estimates * modeled inflation-linked bond with 2% coupon. Calculations are based on the historical dynamic of OFZ yields, indexation to inflation according to the terms of planned OFZ-IN. The expected inflation rate is based on Economy Ministry forecasts and Gazprombank estimates (for Dec.14 Jan.15). 2 Copyright

3 The potential investor base in Russian sovereign inflation-linked papers includes: Pension funds and other institutional investors. In 2015, the inflow to non-state pension funds (NPFs) stands at RUB ~0.5 trln, and contributions have already lent noticeable support to demand on the primary ruble debt market in 2Q15. Bonds that provide a hedge from inflation represent the most natural instrument for funds to invest over a long-term period. Over the past 10 years, the average return on investment of pension money has not covered inflation. Until now, corporate bonds with a coupon rate linked to inflation were the only instrument partially satisfying the goal to obtain a positive real return. However, payment of the principal on such issues is not protected from erosion due to inflation. The structure of the planned issue with principal indexed to inflation ensures maximum protection (principal and coupon payments) against inflation. Inflation-linked bonds secure the highest returns during periods of accelerated CPI growth, which often coincide with periods of monetary policy tightening when traditional bonds as a rule underperform (see Table 2). That said, ILB in theory represents a countercyclical instrument, protecting long-term investment from economic cycle dynamics. Corporate and bank treasuries aiming to preserve the real value of investment, improve the asset structure and hedge from inflation risks. ILB is an attractive tool for investors with inflation-linked liabilities (including indirectly through rates on deposits). Foreign institutional investors. OFZ-IN inclusion into one of the benchmark indexes, in particular the Barclays Emerging Markets Government Inflation-Linked Bonds (EMGILB), could trigger demand on the part of non-residents. Rules allow for inclusion of bonds from an emerging markets country (currently 11 countries included) with a minimum 1 year remaining life. It is important to note that there are no credit rating restrictions for the EMGILB Index. Investors in CIS assets. Inflation rates in Russia and certain CIS states (in particular, Kazakhstan and Azerbaijan) have shown similar dynamics over the past 10 years (see Appendix 5). Inflation-linked OFZs could be used to hedge inflation risks not only in Russia, but also in a number of neighboring countries. Table 2. ILB vs. traditional bonds: choice depends on the stage of the economic cycle ECONOMIC CYCLE INFLATION EXPECTATION AND RISK PREMIUM CENTRAL BANK ACTIONS IMPACT ON BOND MARKET Overheating Low, but rising Start tightening Negative Real Cooling Overshooting Tightening Less negative Real Recession High, but sinking Start softening Positive Nominal Recovery Undershooting Softening Less positive Nominal POSITIVE FOR REAL OR NOMINAL RATES Source: Union Investment 3 Copyright

4 MAY 10 AUG 10 NOV 10 FEB 11 MAY 11 AUG 11 NOV 11 FEB 12 MAY 12 AUG 12 NOV 12 FEB 13 MAY 13 AUG 13 NOV 13 FEB 14 MAY 14 AUG 14 NOV 14 FEB 15 MAY 15 JULY 10, 2015 Inflation-linked bonds protect from ruble devaluation risks In Russia, inflation is tightly correlated with the national currency s exchange rate the correlation coefficient between monthly CPI (YoY) and the USD/RUB exchange rate taken with a two-month time lag stands at 95% for the period since 2012 and above 80% over the past five years. Since 2009, Russia s cumulative inflation has reached 72%, while the USD/RUB exchange rate increased 90% over the same period and the EUR/RUB rate gained 42%. Therefore, inflation-linked OFZs can provide investors with protection not only from inflation risks, but also from ruble devaluation risks. Chart 3. USDRUB exchange rate makes a significant contribution to inflation in Russia Correlation coefficient 83% USDRUB (2M TIME LAG) INFLATION YOY (RHS) Source: Bloomberg, Gazprombank estimates 4 Copyright

5 Bond structure The terms and conditions of inflation-linked OFZs are available at the following link Table 3. Parameters of index-linked bonds offered by the Finance Ministry INDEX-LINKED OFZ Inflation index Time lag for inflation index used for calculations Structure of cash flows Cumulative monthly CPI for goods and services, calculated based on average prices of Bloomberg: RUCPNL Index Three months Capital indexed bond ( Canadian model ) - Fixed coupon rate - Coupon and accrued interest are calculated based on indexed principal value - Indexed principal is paid at redemption Principal floorat redemption 100% Source: The structure is in line with the Capital indexed bond model, which is widely used in both EM and DM (particularly in the US). Cash flows are calculated based on the bond s principal, adjusted for the increase in the selected inflation index since the moment of placement. The inflation data is taken with a three-month time lag. The terms also envisage protection against deflation : the principal floor at redemption is set at 100%. An example of calculations of OFZ-IN cash flows is provided in Appendix 1. Chart 4. OFZ-IN cash flows 5% 4% 3% 2% 1% REAL COUPON RATE 2% 140% 120% 100% 80% 60% 40% 20% 0% 0% DURATION, YEARS COUPONS (REAL) COUPON INDEXATION PRINCIPAL (RHS) INDEXED PART OF PRINCIPAL (RHS) Source: Bloomberg, Gazprombank estimates * cash flows were modeled based on Gazprombank s forecast for inflation and the 5Y OFZ yield 5 Copyright

6 Pricing of inflation-linked OFZs It is assumed that OFZ-IN will be quoted as a percentage of the current (indexed) nominal value in line with the common practice on global ILB markets. These prices, which are usually called real prices, do not include an accumulated inflation part. Yields calculated based on these prices and fixed coupon rates are called real yields according to market practice. The concept of a breakeven inflation rate (average expected inflation on the horizon of the bond s circulation) is commonly used for ILB pricing relative to conventional bonds. In simplified form, a nominal yield of ordinary fixed-coupon bonds could be represented as a function of a real yield and inflation expectations (breakeven): (1 + nominal yield) = (1 + real yield of ILB) x (1 + breakeven inflation) That said, breakeven inflation (BEI) could be further split into: 1) inflation expectations and 2) an inflation risk premium, which should be contained in the yield of the conventional bond. The breakeven rate could be determined more precisely as average inflation through the remaining life of the bond, which equals the present value of ILB cash flows discounted by the nominal yield of a conventional bond to the current price of an ILB. If the ILB parameters (coupon rate, maturity) and the yield of an ordinary bond are known, the BEI is calculated based on an investor s subjective inflation expectations. That said, the observed market BEI could differ from an investor s expectations: if the observed BEI (i.e. the difference between the nominal yield of a conventional bond and the ILB real yield) is above the expected level, a conventional bond looks more attractive for investment. If the observed BEI is below the investor s expectations, ILBs are more appealing relative to conventional bonds. When assessing ILBs, the key issue is how to choose the expected inflation figure. Currently, there are a number of forecasts regarding future inflation which indicates high inflation risks. One of the goals of this new market instrument is to narrow the range of inflation forecasts and become an indicator of market inflation expectations. Official inflation forecasts for 2016 are in the range of 6-7%. However, the CBR is in no rush to lower its key rate at a faster pace; on the contrary, the regulator said that it plans to cut the key rate gradually in light of ongoing inflation risks. Among the factors that prevent the CBR from taking more decisive action, we highlight ongoing high inflation expectations, possible deterioration of global economic conditions, revision of planned increases in regulated prices and tariffs in , and loosening of budgetary policy. Notably, food products prevail in the structure of Russia s CPI, which adds to the volatility of the country s inflation rate (see Appendix 3). In addition, the ruble devaluation effect on prices of imported and domestic goods, as well as introduced trade restrictions, have significantly boosted volatility in inflation starting from We estimate that the contribution of monetary inflation in the overall inflation figure stood at around 30% starting from 2007 and just 10-15% since the beginning of this year (see Appendix 4). As a result, official inflation expectations have yet to gain full support by investors (expressed in purchases of OFZs with yields of ~11% in expectation of further sharp deceleration of inflation to the target level of 4%). The fact that the real inflation rate has often been above the official target also prevents investors from betting on a sharp decline in OFZ yields. For example, in 2012, target inflation stood at 5-6%, while the 6 Copyright

7 actual figure reached 6.6%; in 2013, the target figure (5-6%) was also lower than the actual number (6.5%); and the same trend was seen in 2014 (5.0% vs. 11.4%, respectively). Last year, against the backdrop of the imposition of trade restrictions and the significant impact from ruble devaluation, the real inflation rate diverged from the initial forecast figure by 6.4%. Corporate bonds with coupons linked to the inflation rate, in particular the RZD 32, can serve as an important instrument for assessing the real yield of the placed OFZ-IN bond issue. According to our estimates, current quotes of the abovementioned issue (at around 102.3% of par value) and the positioning of the Russian Railways conventional bond curve (spread of ~100 bps over OFZs) assume implied long-term inflation of % depending on the model chosen to determine the expected inflation (see Model risks section). Given these figures, the real yield of the 8Y paper could amount to % annually, while expected (breakeven) inflation could reach %. This level of real yields is in line with similar rates in such countries as South Africa (~1.7% for 8-10Y), Turkey and Mexico ( % for 8-10Y), as well as Russia s potential GDP growth rate in the medium term (1-2%). Model risks The calculation of OFZ-IN price/yield depends not only on inflationary expectations and current OFZ price parameters, but also only the technical aspects of determining future CPI indicators used for indexation of cash flows. In particular, the model may use inflation forecasts in terms of year on year (a cumulative indicator for 12 months: for example, cumulative inflation for July 2014 June 2015 totals 15.3%). In this case, CPI will be derived as CPI 12 months ago, adjusted for cumulative annual inflation over the calculation period. This approach corresponds to the conditions of corporate bonds with coupons tied to inflation (RZD 32, FGC 22). The drawback of this approach is the volatile dynamic of CPI in the future due to shifts of past years inflationary shocks (see Chart 5). A second approach assumes that inflationary expectations represent an annualized month on month figure. In this case, CPI is derived as the figure of the previous month, adjusted for monthly inflation. Monthly inflation is calculated according to the formula: Inflation month/month = ( Inflation for the month/month, annualized + 1) ^(1/12). The use of monthly inflation, in our view, is more appropriate for the OFZ-IN model as the CPI index is calculated on a monthly basis. At the same time, we note the fact that annualized monthly inflation in Russia has historically been much more volatile than cumulative inflation over the past 12 months, and the two indicators may differ noticeably. The results of our valuation of OFZ-IN according to the two different approaches to determining forecast inflation are presented in Table 4. In case of use of a market estimate of inflationary expectations, the OFZ-IN price/yield will also depend on the price quality for market instruments used to derive implied inflation (RZD 32). Based on our estimates, each 1 pp change in the price of the RZD 32 leads to a change in implied inflation of up to 30 bps and a change in real yield of up to 25 bps. 7 Copyright

8 JULY 10, 2015 Table 4. Pricing for 8Y OFZ-IN depending on the inflation model INFLATION OUTLOOK MOM INFLATION OUTLOOK YOY Coupon 2% Date of placement * July 15, 2015 Implied inflation ** 8.2% 9.1% Price 96.9% 100.6% Real yield 2.4% 1.9% Breakeven 8.7% 9.2% Chart 5. CPI* dynamic depending on the determination of expected inflation 1,200 1,100 1, CPI, INFLATION OUTLOOK YOY Source: Gazprombank estimates * taken for illustrative purposes only. According to the Finance Ministry, the placement could take place before the end of July, 2015 CPI, INFLATION OUTLOOK MOM ** inflation implied by the RZD 32 bond priced at 102.3% and the conventional RZD curve Chart 6. 12M inflation (YoY) distribution, H15 30% 25% 20% 15% 10% 5% 0% Frequency 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% SINCE 2001 OVER PAST 5 YEARS Inflation Source: Gazprombank estimates * cumulative CPI, calculated to average prices of 2000 Source: Gazprombank estimates According to analyst forecasts provided by Bloomberg, inflation expectations for 2016 range from 5-10% YoY, which covers both the official forecast and implied inflation priced in the traded instruments (corporate bonds with an inflation-linked coupon). In our opinion, the fundamental factor that stands behind the variance in inflation assessments could be the significant inflation risk premium in Russia. The risks that inflation could deviate from the official forecast stem from risks related to the ruble exchange rate dynamic and CBR monetary policy. Notably, the appearance of a new market indicator of inflation expectations (OFZ IN) should help constrain the variance of inflation estimates provided by market participants, as well as provide the CBR with an important tool to assess the market s response to the official inflation targets. In the future, this should help reduce the inflation risk premium, which could be supportive for conventional OFZs. We also assessed inflation-linked OFZs based on different inflation expectations, current OFZ yields and various estimates of the bond s coupon rates. The results of our assessment are presented in Chart 8. 8 Copyright

9 Chart 7. OFZ-IN real yield, breakeven and conventional OFZ nominal yield Chart 8. Sensitivity analysis based on inflation outlook YTW, % OFZ YIELD CURVE inflation risk premium expected inflation real yield breakeven MATURITY, YEARS Source: Gazprombank estimates PRICE MEDIUM-TERM FORECASTS* 50 3% 4% 5% 6% 7% 8% 9% 10% 11% INFLATION EXPECTATIONS, INCLUDING INFLATION RISK PREMIUM COUPON 1% COUPON 2% COUPON 3% COUPON 4% Source: Gazprombank estimates * According to Bloomberg, forecast for 2016 Taxation of OFZ IN For legal entities (Russian residents), the tax regime governing OFZ IN will differ from that for conventional OFZs (see Table 5). In particular, the inflation-indexed part of principal will be taxed at a rate of 15%, which is similar to the rate at which coupon yield is taxed. Capital gains on conventional OFZs are taxed at a rate of 20%. In practice, the relative tax appeal of OFZ IN compared with conventional OFZs will be determined by a number of factors, including the absolute inflation level and its dynamic, as well as the parameters of conventional OFZs (coupon rate, purchase or offer price). For example, in the case of inflation deceleration during the OFZ IN circulation, the tax advantages of the former compared with traditional OFZs could increase, and on the contrary decrease should CPI accelerate. Table 5. Tax rates on return from OFZ IN for different investor categories CAPITAL GAIN COUPON INDEXED PART OF PRINCIPAL Legal entities 20% 15% 15% Residents Individuals 13% 0% 0% NPFs* 20% 15% 15% Non-residents Exempted from tax Source: Finance Ministry * tax collected when return exceeds the CBR refinancing rate 9 Copyright

10 Appendix 1. OFZ-IN: practical example of cash flows and index ratio calculation Input data: Date of placement of OFZ-IN: July 15, 2015 (not a defined placement date, taken for illustrative purposes only) Par value at time of placement: RUB 1000 Coupon rate: 2%. Calculation of indexed principal: P i = 1000 х I (where I = current inflation index / base inflation index on date of issue). At issue date I = 1, principal = RUB 1000 Calculation of inflation index: Current inflation index = CPI 4 months ago + ( CPI 4 months ago CPI 3 months ago ) ( current day 1) / number of days in the current month. CPI is published on the State Statistics Service website, and is also available in Bloomberg <RUCPNL Index>. Example: Base inflation index = CPI for March 2015 (month number: 7 4 = 3) + (CPI for April 2015 CPI for March 2015) х (15 1 / 31) = ( ) х (14/31) = (accurate to within five decimal places). Inflation index as of July 16 = ( ) х (15/31) = I = / = N i = 1000 x = RUB The Finance Ministry publishes data on the par value of bonds for each calendar day of the month on its official website no later than two working days prior to the beginning of this month. Calculation of accrued interest: Example: CCI = Н i х Coupon rate х Number of days passed within coupon period / 365. AI July 16 = х 1 / 365 х 2% = 5.5 kopecks. Coupon income is calculated in the same way. 10 Copyright

11 Appendix 2. Breakevens expected inflation and real yields longterm growth rates relation: international experience Our analysis showed that breakeven inflation in DM and EM countries precisely reflects forecasts of average annual inflation over the horizon of the bond s circulation. We used the IMF s average 5-year inflation forecast. Forecasts provided by experts seem to have an impact on investor expectations, and conversely, economists may use market data in their models. As a result, market and expert estimates converge. In turn, real yields show a positive relationship with long-term economic growth rates (see Diagram 10). In theory, high GDP growth rates heighten expectations of a future increase in real income, which leads to a higher propensity to consume. Consequently, investors want to see a sufficient level of rates capable of persuading them to switch from current consumption in favor of savings. At the same time, real yields depend on other factors that determine the dynamics of all financial assets. In particular, real yields of 5Y ILBs in DM countries are currently in the area of zero and negative values due to QE and accommodative monetary policy. Chart 9. Correlation of breakeven rate with expected inflation 8% 7% 6% 5% 4% 3% 2% BREAKEVEN FOR 5Y ILB, % UK US MEXICO R² = TURKEY SOUTH AFRICA BRAZIL 1% GERMANY ITALY 0% 0% 1% 2% 3% 4% 5% 6% 7% AVERAGE EXPECTED INFLATION, 5Y Source: Bloomberg, IMF Chart 10. Correlation of real yield with long-term growth rates 7% 6% 5% 4% 3% REAL YIELD OF 5Y ILB, % BRAZIL R² = TURKEY 2% 1% SOUTH AFRICA MEXICO ITALY US 0% 0.0% 0.5% 1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% -1% GERMANY -2% UK GDP GROWTH, 2020 Source: Bloomberg, IMF 11 Copyright

12 Appendix 3. CPI baskets in Russia and other EM countries: a large share of food commodities causes higher volatility. A comparison of consumer baskets used to calculate inflation in developed and developing countries indicates a relatively high proportion of food in the structure of the latter (25-30% vs. 15% in DM countries), which is reflected in more volatile inflation dynamics. Chart 11. Consumer basket in Russia Chart 12. Consumer basket in Brazil 17% 37% 39% 25% 9% 37% FOOD PRODUCTS NON-FOOD PRODUCTS HOUSING SERVICES AND UTILITIES OTHER SERVICES Chart 13. Consumer basket in Mexico FOOD PRODUCTS 15% HOUSING SERVICES AND UTILITIES Chart 14. Consumer basket in Turkey 21% NON-FOOD PRODUCTS OTHER SERVICES 26% 23% 35% 29% 28% 22% 16% 20% FOOD PRODUCTS HOUSING SERVICES AND UTILITIES NON-FOOD PRODUCTS OTHER SERVICES FOOD PRODUCTS HOUSING SERVICES AND UTILITIES NON-FOOD PRODUCTS OTHER SERVICES Chart 15. Consumer basket in the US Chart 16. Consumer basket in the UK 30% 14% 37% 15% 20% 37% FOOD PRODUCTS HOUSING SERVICES AND UTILITIES NON-FOOD PRODUCTS OTHER SERVICES FOOD PRODUCTS 13% HOUSING SERVICES AND UTILITIES 35% NON-FOOD PRODUCTS OTHER SERVICES Source: State Statistics Service, Banco Central do Brasil, Banco de Mexico, ENEGI, Turkstat, US Department of Labor, Office of National Statistics, OECD, Gazprombank estimates 12 Copyright

13 FEB 04 JUL 04 DEC 04 MAY 05 OCT 05 MAR 06 AUG 06 JAN 07 JUN 07 NOV 07 APR 08 SEP 08 FEB 09 JUL 09 DEC 09 MAY 10 OCT 10 MAR 11 AUG 11 JAN 12 JUN 12 NOV 12 APR 13 SEP 13 FEB 14 JUL 14 DEC 14 MAY 15 JULY 10, 2015 Russia has the highest share of food (37%) among the considered countries, which. could become one of the fundamental conditions for higher inflation risks compared with EM peers. According to this logic, inflation-linked OFZs could potentially occupy a certain market niche, whose relative size will be no less than in countries with a more stable inflation rate (Brazil 37% of the internal government bond market, Mexico 13%). Chart 17. Among EM peers, Russia has the most volatile inflation and the highest share of food in a basket of consumer goods % RUSSIA MEXICO TURKEY BRAZIL 13 Copyright

14 JUN 07 SEP 07 DEC 07 MAR 08 JUN 08 SEP 08 DEC 08 MAR 09 JUN 09 SEP 09 DEC 09 MAR 10 JUN 10 SEP 10 DEC 10 MAR 11 JUN 11 SEP 11 DEC 11 MAR 12 JUN 12 SEP 12 DEC 12 MAR 13 JUN 13 SEP 13 DEC 13 MAR 14 JUN 14 SEP 14 DEC 14 MAR 15 JUN 15 SEP 15 DEC 15 MAR 16 JUN 16 SEP 16 DEC 16 JULY 10, 2015 Appendix 4. Factor analysis of inflation in Russia. As shown below in our factor analysis of inflation in Russia since 2007, historically the greatest contributions to inflation have been made by growth in tariffs, monetary inflation and inflationary expectations, as well as the transfer effect from ruble devaluation on import and domestic prices. The last factor has accounted for most of the spike of inflation seen since December In our base assumptions, this factor will be exhausted by December A second extraordinary factor (embargo) will also cease to play a role by September, as it is overall already reflected in domestic prices. Chart 18. Breakdown of actual and forecast inflation (by end 2016) by main factors STRUCTURAL INERTIAL COMPONENT MONETARY INFLATION AND INFLATION EXPECTATIONS EFFECT FROM EMBARGO INCREASE IN TARIFFS AND REGULATED PRICES GLOBAL FOOD PRICES RUBLE EXCHANGE RATE Source: Gazprombank estimates 14 Copyright

15 DEC 04 MAY 05 OCT 05 MAR 06 AUG 06 JAN 07 JUN 07 NOV 07 APR 08 SEP 08 FEB 09 JUL 09 DEC 09 MAY 10 OCT 10 MAR 11 AUG 11 JAN 12 JUN 12 NOV 12 APR 13 SEP 13 FEB 14 JUL 14 DEC 14 MAY 15 JULY 10, 2015 Appendix 5. Advanced use of ILB inflation risks in FSU? Over the past decade, cumulative inflation rates in FSU states Russia, Kazakhstan and Azerbaijan have shown similar dynamics. The correlation coefficient between the relevant indicators of Russia and Kazakhstan stands at 99.8%, and between Russia and Azerbaijan at 93.9%. This could be explained by the similar oil and gas export orientation, and the interconnection between the economy and external trade. The shares of Russia in the external trade turnover of Kazakhstan and Azerbaijan stand at 23% and 8%, respectively. A similar CPI basket facilitates the near-term inflation dynamic. For example, in Russia and Kazakhstan food accounts for shares of 37.3% and 36.4%, non-food goods for 37.1% and 32.5%, and services for 25.6% and 31.4%, respectively. Based on the near-term inflation dynamic in these countries, OFZs with an indexed par value may, with certain reservations, be used to hedge the risk of unforeseen inflation not only in Russia, but also in neighboring countries. We note that investors should separately hedge currency risk. Chart 19. Cumulative inflation in CIS countries (2005=100) RUSSIA KAZAKHSTAN AZERBAIJAN Source: World Bank, Bloomberg The dynamic in currencies of select countries also shows a similar directionality, although the ruble is characterized by higher volatility due to an initially more flexible exchange rate policy in Russia and the subsequent transition to a free-floating regime. Synchronization of the exchange rate dynamic in these countries could facilitate a convergence of monetary policies. Kazakhstan unveiled plans for a gradual transition, similar to Russia, to inflation targeting by At the beginning of 2015, Azerbaijan took the first steps toward inflation targeting: the country s central bank removed the national currency s peg to the dollar in favor of a bi-currency basket and announced a future inflation target of 5-7%. A shift to inflation targeting requires a free-floating exchange rate regime in neighboring countries, which could narrow the differences in the national exchange rate dynamics and inflation, induced by volatility on currency markets. 15 Copyright

16 MAR 09 JUN 09 SEP 09 DEC 09 MAR 10 JUN 10 SEP 10 DEC 10 MAR 11 JUN 11 SEP 11 DEC 11 MAR 12 JUN 12 SEP 12 DEC 12 MAR 13 JUN 13 SEP 13 DEC 13 MAR 14 JUN 14 SEP 14 DEC 14 MAR 15 JUN 15 JULY 10, 2015 Chart 20. RUB, KZT and AZN dynamics vs. USD since the start of RUB KZT AZN Source: Bloomberg, Gazprombank estimates 16 Copyright

17 HQ: 16/1 Nametkina St., Moscow , Russia. Office: 7 Koroviy val St. +7 (495) EQUITY SALES +7 (495) FIXED INCOME SALES +7 (495) EQUITY TRADING +7 (495) FIXED INCOME TRADING +7 (499) Copyright All rights reserved This report has been prepared by the analysts of (hereinafter Gazprombank) and is based on information obtained from public sources believed to be reliable, but is not guaranteed as necessarily being accurate. With the exception of information directly pertaining to Gazprombank, the latter shall not be liable for the accuracy or completeness of any information shown herein. All opinions and judgments herein represent solely analysts personal opinion regarding the events and situations described and analyzed in this report. They should not be regarded as Gazprombank s position and are subject to change without notice, also in connection with new corporate or market events that may transpire. Gazprombank shall be under no obligation to update, amend this report or otherwise notify anyone of any such changes. The financial instruments mentioned herein may be unsuitable for certain categories of investors. This report should not be the only basis used when adopting an investment decision. Investors should make investment decisions at their own discretion, inviting independent consultants, if necessary, for their specific interests and objectives. The authors shall not be liable for any actions resulting from the use of this report. Any information contained herein or in the appendices hereto shall not to be construed as a solicitation or an offer to buy or sell any securities or advertisement, unless otherwise expressly stated herein or in the appendices hereto. Gazprombank should in no way be viewed as soliciting, facilitating, brokering or causing any persons to invest in any instrument or otherwise engage in transactions that may be prohibited to those persons under applicable laws and regulations. Investors should independently evaluate whether any investment transactions undertaken after reviewing Gazprombank research materials are permitted under any applicable economic sanctions laws and regulations or other laws applicable to their investing activities.